EduRABSA_SLM
Collection
Small multi-task generative LLMs, fine-tuned for education review ABSA (aspect-based sentiment analysis) tasks using the EduRABSA dataset.
•
9 items
•
Updated
Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: CastError
Message: Couldn't cast
indices: uint64
-- schema metadata --
huggingface: '{"info": {"features": {"indices": {"dtype": "uint64", "_typ' + 50
to
{'id': Value('string'), 'task_type': Value('string'), 'entry_id': Value('string'), 'original_id': Value('string'), 'input': Value('string'), 'prompt': Value('string'), 'entry_list': List(Value('string')), 'output': List(Value('string')), 'text': Value('string'), 'type': Value('string')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1815, in _prepare_split_single
for _, table in generator:
^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/arrow/arrow.py", line 76, in _generate_tables
yield f"{file_idx}_{batch_idx}", self._cast_table(pa_table)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/arrow/arrow.py", line 59, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2272, in table_cast
return cast_table_to_schema(table, schema)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2218, in cast_table_to_schema
raise CastError(
datasets.table.CastError: Couldn't cast
indices: uint64
-- schema metadata --
huggingface: '{"info": {"features": {"indices": {"dtype": "uint64", "_typ' + 50
to
{'id': Value('string'), 'task_type': Value('string'), 'entry_id': Value('string'), 'original_id': Value('string'), 'input': Value('string'), 'prompt': Value('string'), 'entry_list': List(Value('string')), 'output': List(Value('string')), 'text': Value('string'), 'type': Value('string')}
because column names don't match
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1455, in compute_config_parquet_and_info_response
parquet_operations = convert_to_parquet(builder)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1054, in convert_to_parquet
builder.download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 894, in download_and_prepare
self._download_and_prepare(
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 970, in _download_and_prepare
self._prepare_split(split_generator, **prepare_split_kwargs)
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1702, in _prepare_split
for job_id, done, content in self._prepare_split_single(
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.12/site-packages/datasets/builder.py", line 1858, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
id
string | task_type
string | entry_id
string | original_id
string | input
string | prompt
string | entry_list
list | output
list | text
string | type
string |
|---|---|---|---|---|---|---|---|---|---|
ASTE_10
|
ASTE
|
10
|
2_8704_481
|
No reason for this course to exist
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>null</asp><opn>No reason for this course to exist</opn><sen>negative</sen>"
] |
[
"<asp>null</asp><opn>No reason for this course to exist</opn><sen>negative</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```No reason for this course to exist```
|
course review
|
ASTE_23
|
ASTE
|
23
|
1_89_2548_10676
|
He is the Man. Best Teacher i have had. Really teaches you, and will help you. I took him for my first college writing class, has made the rest of my college career way easier. Cant say enough good things. Take his class, you wont be disappointed. You talk movies and have a good time while actually learning real skills
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>He</asp><opn>has made the rest of my college career way easier</opn><sen>positive</sen>",
"<asp>He</asp><opn>is the Man</opn><sen>positive</sen>",
"<asp>He</asp><opn>Best Teacher i have had</opn><sen>positive</sen>",
"<asp>He</asp><opn>Really teaches you</opn><sen>positive</sen>",
"<asp>He</asp><opn>will help you</opn><sen>positive</sen>",
"<asp>He</asp><opn>Cant say enough good things</opn><sen>positive</sen>",
"<asp>his class</asp><opn>You talk movies and have a good time while actually learning real skills</opn><sen>positive</sen>",
"<asp>his class</asp><opn>you wont be disappointed</opn><sen>positive</sen>"
] |
[
"<asp>He</asp><opn>has made the rest of my college career way easier</opn><sen>positive</sen>",
"<asp>He</asp><opn>is the Man</opn><sen>positive</sen>",
"<asp>He</asp><opn>Best Teacher i have had</opn><sen>positive</sen>",
"<asp>He</asp><opn>Really teaches you</opn><sen>positive</sen>",
"<asp>He</asp><opn>will help you</opn><sen>positive</sen>",
"<asp>He</asp><opn>Cant say enough good things</opn><sen>positive</sen>",
"<asp>his class</asp><opn>You talk movies and have a good time while actually learning real skills</opn><sen>positive</sen>",
"<asp>his class</asp><opn>you wont be disappointed</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```He is the Man. Best Teacher i have had. Really teaches you, and will help you. I took him for my first college writing class, has made the rest of my college career way easier. Cant say enough good things. Take his class, you wont be disappointed. You talk movies and have a good time while actually learning real skills```
|
teacher review
|
ASTE_59
|
ASTE
|
59
|
3_234_502_198
|
I regret coming to this university very much. It lacks sufficient support and usually does not provide advice, but only follows the rules and procedures. Students have no chance to be guided to visit the campus in a deeper way. Even so, tuition fees have not been reduced.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>this university</asp><opn>regret coming</opn><sen>negative</sen>",
"<asp>support</asp><opn>lacks sufficient</opn><sen>negative</sen>",
"<asp>advice</asp><opn>usually does not provide</opn><sen>negative</sen>",
"<asp>rules and procedures</asp><opn>only follows</opn><sen>negative</sen>",
"<asp>visit the campus</asp><opn>no chance to be guided</opn><sen>negative</sen>",
"<asp>tuition fees</asp><opn>have not been reduced</opn><sen>negative</sen>"
] |
[
"<asp>this university</asp><opn>regret coming</opn><sen>negative</sen>",
"<asp>support</asp><opn>lacks sufficient</opn><sen>negative</sen>",
"<asp>advice</asp><opn>usually does not provide</opn><sen>negative</sen>",
"<asp>rules and procedures</asp><opn>only follows</opn><sen>negative</sen>",
"<asp>visit the campus</asp><opn>no chance to be guided</opn><sen>negative</sen>",
"<asp>tuition fees</asp><opn>have not been reduced</opn><sen>negative</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```I regret coming to this university very much. It lacks sufficient support and usually does not provide advice, but only follows the rules and procedures. Students have no chance to be guided to visit the campus in a deeper way. Even so, tuition fees have not been reduced.```
|
university review
|
ASTE_91
|
ASTE
|
91
|
1_206_1040_14559
|
A highlighter is your best friend in this class, everything is directly taken from the power points he uses or from the book; mainly the book. READ THE CHAPTERS, otherwise you're in for a hell of a time. His tone of voice is dull as watching paint dry, so if you like to engage in class there will be plenty of time to do so, as he encourages input.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>null</asp><opn>everything is directly taken from the power points he uses or from the book</opn><sen>neutral</sen>",
"<asp>His tone of voice</asp><opn>dull as watching paint dry</opn><sen>negative</sen>",
"<asp>he</asp><opn>encourages input</opn><sen>positive</sen>"
] |
[
"<asp>null</asp><opn>everything is directly taken from the power points he uses or from the book</opn><sen>neutral</sen>",
"<asp>His tone of voice</asp><opn>dull as watching paint dry</opn><sen>negative</sen>",
"<asp>he</asp><opn>encourages input</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```A highlighter is your best friend in this class, everything is directly taken from the power points he uses or from the book; mainly the book. READ THE CHAPTERS, otherwise you're in for a hell of a time. His tone of voice is dull as watching paint dry, so if you like to engage in class there will be plenty of time to do so, as he encourages input.```
|
teacher review
|
ASTE_95
|
ASTE
|
95
|
1_45_1105_6306
|
She really knows her material and is a good teacher but can be a little rude at times but still nice.. studying everyday is a must if you want to do good and the homework she assigns. She's willing to help her students if they need help. Her tests are a lil tricky make sure you spend a couple days in advance studying overall shes an okay teacher
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>teacher</asp><opn>really knows her material</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>good</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>willing to help</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>a little rude</opn><sen>negative</sen>",
"<asp>teacher</asp><opn>still nice</opn><sen>positive</sen>",
"<asp>Her tests</asp><opn>a lil tricky</opn><sen>negative</sen>",
"<asp>teacher</asp><opn>okay</opn><sen>neutral</sen>"
] |
[
"<asp>teacher</asp><opn>really knows her material</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>good</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>willing to help</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>a little rude</opn><sen>negative</sen>",
"<asp>teacher</asp><opn>still nice</opn><sen>positive</sen>",
"<asp>Her tests</asp><opn>a lil tricky</opn><sen>negative</sen>",
"<asp>teacher</asp><opn>okay</opn><sen>neutral</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```She really knows her material and is a good teacher but can be a little rude at times but still nice.. studying everyday is a must if you want to do good and the homework she assigns. She's willing to help her students if they need help. Her tests are a lil tricky make sure you spend a couple days in advance studying overall shes an okay teacher```
|
teacher review
|
ASTE_503
|
ASTE
|
503
|
3_99_146_44
|
Actually the stuff in the business school are really welcoming. Especially Bill Peng
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>stuff in the business school</asp><opn>welcoming</opn><sen>positive</sen>",
"<asp>Bill Peng</asp><opn>welcoming</opn><sen>positive</sen>"
] |
[
"<asp>stuff in the business school</asp><opn>welcoming</opn><sen>positive</sen>",
"<asp>Bill Peng</asp><opn>welcoming</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Actually the stuff in the business school are really welcoming. Especially Bill Peng```
|
university review
|
ASTE_510
|
ASTE
|
510
|
2_3791_392
|
Material was pure memorization and was quite easy to grasp once you read the textbook
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>Material</asp><opn>quite easy to grasp once you read the textbook</opn><sen>positive</sen>",
"<asp>Material</asp><opn>pure memorization</opn><sen>neutral</sen>"
] |
[
"<asp>Material</asp><opn>quite easy to grasp once you read the textbook</opn><sen>positive</sen>",
"<asp>Material</asp><opn>pure memorization</opn><sen>neutral</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Material was pure memorization and was quite easy to grasp once you read the textbook```
|
course review
|
ASTE_529
|
ASTE
|
529
|
2_12193_814
|
This course honestly kinda sucked. I took it online. Although it may have been kinda useful in the sense that you got to learn about Asian history, there wasn't many ways of delivering content. There were a TON of readings, and the quizzes were based off of them. I had to write 3 Asian movie reviews which were the best part about the course.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>course</asp><opn>wasn't many ways of delivering content</opn><sen>negative</sen>",
"<asp>course</asp><opn>honestly kinda sucked</opn><sen>negative</sen>",
"<asp>course</asp><opn>kinda useful</opn><sen>neutral</sen>",
"<asp>readings</asp><opn>a TON of</opn><sen>negative</sen>",
"<asp>Asian movie reviews</asp><opn>the best part about the course</opn><sen>positive</sen>"
] |
[
"<asp>course</asp><opn>wasn't many ways of delivering content</opn><sen>negative</sen>",
"<asp>course</asp><opn>honestly kinda sucked</opn><sen>negative</sen>",
"<asp>course</asp><opn>kinda useful</opn><sen>neutral</sen>",
"<asp>readings</asp><opn>a TON of</opn><sen>negative</sen>",
"<asp>Asian movie reviews</asp><opn>the best part about the course</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```This course honestly kinda sucked. I took it online. Although it may have been kinda useful in the sense that you got to learn about Asian history, there wasn't many ways of delivering content. There were a TON of readings, and the quizzes were based off of them. I had to write 3 Asian movie reviews which were the best part about the course.```
|
course review
|
ASTE_554
|
ASTE
|
554
|
2_13305_629
|
a nice break from stem. wasnt just random old " classics " you write your own stuff!!
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>null</asp><opn>nice break from stem</opn><sen>positive</sen>",
"<asp>null</asp><opn>wasnt just random old \" classics \"</opn><sen>positive</sen>"
] |
[
"<asp>null</asp><opn>nice break from stem</opn><sen>positive</sen>",
"<asp>null</asp><opn>wasnt just random old \" classics \"</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```a nice break from stem. wasnt just random old " classics " you write your own stuff!!```
|
course review
|
ASTE_589
|
ASTE
|
589
|
3_29_38_86
|
I particularly like this school, especially the career zone service. The teachers are also very friendly.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>this school</asp><opn>particularly like</opn><sen>positive</sen>",
"<asp>career zone service</asp><opn>particularly like</opn><sen>positive</sen>",
"<asp>teachers</asp><opn>very friendly</opn><sen>positive</sen>"
] |
[
"<asp>this school</asp><opn>particularly like</opn><sen>positive</sen>",
"<asp>career zone service</asp><opn>particularly like</opn><sen>positive</sen>",
"<asp>teachers</asp><opn>very friendly</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```I particularly like this school, especially the career zone service. The teachers are also very friendly.```
|
university review
|
ASTE_594
|
ASTE
|
594
|
1_190_631_16994
|
Demaree is an awesome nutritiom teacher. She is maily does slides in her class, and if you take all of your notes off of them you are guarunteed to do well on her tests. She is really nice and really helpful.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>Demaree</asp><opn>awesome</opn><sen>positive</sen>",
"<asp>slides</asp><opn>guarunteed to do well on her tests</opn><sen>positive</sen>",
"<asp>Demaree</asp><opn>really nice</opn><sen>positive</sen>",
"<asp>Demaree</asp><opn>really helpful</opn><sen>positive</sen>"
] |
[
"<asp>Demaree</asp><opn>awesome</opn><sen>positive</sen>",
"<asp>slides</asp><opn>guarunteed to do well on her tests</opn><sen>positive</sen>",
"<asp>Demaree</asp><opn>really nice</opn><sen>positive</sen>",
"<asp>Demaree</asp><opn>really helpful</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Demaree is an awesome nutritiom teacher. She is maily does slides in her class, and if you take all of your notes off of them you are guarunteed to do well on her tests. She is really nice and really helpful.```
|
teacher review
|
ASTE_601
|
ASTE
|
601
|
2_13846_1643
|
Super interesting. Dr. Cyr posts content in podcast form with notes to follow along. Final essay as well as several shorter journal assignments.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>null</asp><opn>Super interesting</opn><sen>positive</sen>"
] |
[
"<asp>null</asp><opn>Super interesting</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Super interesting. Dr. Cyr posts content in podcast form with notes to follow along. Final essay as well as several shorter journal assignments.```
|
course review
|
ASTE_638
|
ASTE
|
638
|
1_89_1151_16167
|
Very great teacher. Encourages personal thinking. Lots of work goes into earning the grade but I loved the books we read and the support to think about everything and not just take it at face value. A REAL TEACHER! Teaching us to think for ourselves.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>teacher</asp><opn>Very great</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>Teaching us to think for ourselves</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>Encourages personal thinking</opn><sen>positive</sen>",
"<asp>earning the grade</asp><opn>Lots of work</opn><sen>negative</sen>",
"<asp>books we read</asp><opn>loved</opn><sen>positive</sen>",
"<asp>support to think about everything</asp><opn>loved</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>A REAL TEACHER</opn><sen>positive</sen>"
] |
[
"<asp>teacher</asp><opn>Very great</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>Teaching us to think for ourselves</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>Encourages personal thinking</opn><sen>positive</sen>",
"<asp>earning the grade</asp><opn>Lots of work</opn><sen>negative</sen>",
"<asp>books we read</asp><opn>loved</opn><sen>positive</sen>",
"<asp>support to think about everything</asp><opn>loved</opn><sen>positive</sen>",
"<asp>teacher</asp><opn>A REAL TEACHER</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Very great teacher. Encourages personal thinking. Lots of work goes into earning the grade but I loved the books we read and the support to think about everything and not just take it at face value. A REAL TEACHER! Teaching us to think for ourselves.```
|
teacher review
|
ASTE_669
|
ASTE
|
669
|
3_504_499_219
|
Everything about UoE oozes modernity , confidence and pride. The new buildings demonstrate Exeter's advancement into the future and its desire to provide its students with everything they need for a degree in the new 21st century. Societies both academic and vocational are well-run, charismatic and thoughtful, encouraging everyone to participate and have fun. Extra-curricular activities such as help with careers is easy to find , use and learn from, providing vast resources to those who truly need it. In essence, UoE has set the standards for pedigree in the university world.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>UoE</asp><opn>oozes modernity , confidence and pride</opn><sen>positive</sen>",
"<asp>new buildings</asp><opn>demonstrate Exeter's advancement into the future and its desire to provide its students with everything they need</opn><sen>positive</sen>",
"<asp>Societies</asp><opn>encouraging everyone to participate and have fun</opn><sen>positive</sen>",
"<asp>Societies</asp><opn>well-run</opn><sen>positive</sen>",
"<asp>Societies</asp><opn>charismatic and thoughtful</opn><sen>positive</sen>",
"<asp>Extra-curricular activities</asp><opn>easy to find , use and learn from</opn><sen>positive</sen>",
"<asp>resources</asp><opn>vast</opn><sen>positive</sen>",
"<asp>UoE</asp><opn>set the standards for pedigree in the university world</opn><sen>positive</sen>"
] |
[
"<asp>UoE</asp><opn>oozes modernity , confidence and pride</opn><sen>positive</sen>",
"<asp>new buildings</asp><opn>demonstrate Exeter's advancement into the future and its desire to provide its students with everything they need</opn><sen>positive</sen>",
"<asp>Societies</asp><opn>encouraging everyone to participate and have fun</opn><sen>positive</sen>",
"<asp>Societies</asp><opn>well-run</opn><sen>positive</sen>",
"<asp>Societies</asp><opn>charismatic and thoughtful</opn><sen>positive</sen>",
"<asp>Extra-curricular activities</asp><opn>easy to find , use and learn from</opn><sen>positive</sen>",
"<asp>resources</asp><opn>vast</opn><sen>positive</sen>",
"<asp>UoE</asp><opn>set the standards for pedigree in the university world</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Everything about UoE oozes modernity , confidence and pride. The new buildings demonstrate Exeter's advancement into the future and its desire to provide its students with everything they need for a degree in the new 21st century. Societies both academic and vocational are well-run, charismatic and thoughtful, encouraging everyone to participate and have fun. Extra-curricular activities such as help with careers is easy to find , use and learn from, providing vast resources to those who truly need it. In essence, UoE has set the standards for pedigree in the university world.```
|
university review
|
ASTE_670
|
ASTE
|
670
|
3_372_173_239
|
Really good uni- one of the top 10
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>uni</asp><opn>Really good</opn><sen>positive</sen>"
] |
[
"<asp>uni</asp><opn>Really good</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Really good uni- one of the top 10```
|
university review
|
ASTE_694
|
ASTE
|
694
|
2_10923_1183
|
Can be easy or super difficult depending on profs.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>null</asp><opn>depending on profs</opn><sen>neutral</sen>"
] |
[
"<asp>null</asp><opn>depending on profs</opn><sen>neutral</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Can be easy or super difficult depending on profs.```
|
course review
|
ASTE_1018
|
ASTE
|
1018
|
3_499_527_219
|
I've had the best 3 years of my life at Exeter. I love it so much I'm going back next year to do a masters!
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>Exeter</asp><opn>the best 3 years of my life</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>going back next year</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>I love it so much</opn><sen>positive</sen>"
] |
[
"<asp>Exeter</asp><opn>the best 3 years of my life</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>going back next year</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>I love it so much</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```I've had the best 3 years of my life at Exeter. I love it so much I'm going back next year to do a masters!```
|
university review
|
ASTE_1031
|
ASTE
|
1031
|
2_3928_200
|
Please just go to the writing center to get your lab proof read. And make sure to put a citation even if it is a simple scientific fact.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[] |
[] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Please just go to the writing center to get your lab proof read. And make sure to put a citation even if it is a simple scientific fact.```
|
course review
|
ASTE_1035
|
ASTE
|
1035
|
3_134_143_83
|
Great University. The Cornwall campus is absolutely stunning, as beautiful as the Duchy.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>University</asp><opn>Great</opn><sen>positive</sen>",
"<asp>Cornwall campus</asp><opn>beautiful</opn><sen>positive</sen>",
"<asp>Cornwall campus</asp><opn>absolutely stunning</opn><sen>positive</sen>"
] |
[
"<asp>University</asp><opn>Great</opn><sen>positive</sen>",
"<asp>Cornwall campus</asp><opn>beautiful</opn><sen>positive</sen>",
"<asp>Cornwall campus</asp><opn>absolutely stunning</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Great University. The Cornwall campus is absolutely stunning, as beautiful as the Duchy.```
|
university review
|
ASTE_1045
|
ASTE
|
1045
|
3_393_55_287
|
I have just completed my first year at Exeter University and I absolutely loved it. Campus / Facilities: I stayed in the St. Germans( sometimes called just Rowe house but they're not the same!) and literally could not have asked for better. I heard that they're going to tear St Germans down to build more blocks like Lafrowda:( i guess its not cost-efficient to have 36 ppl in a block, with 6 ppl per flat with double beds, en-suite, big kitchen+ a nice living room area w bean bags in each flat! But it was nice for us at least so get in while you can. Gym is also new and super fresh, it can get quite busy but there are a lot of different cardio machines and stuff so its usually fine. actual campus is also nice and modern. Club / societies: there is smt for everyone, so def get involved in at least one or two, makes the year a lot of fun and you always have events to look forward to throughout the year, exeter is quite a small town after all so you need to spice your life up a bit!
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>first year at Exeter University</asp><opn>absolutely loved it</opn><sen>positive</sen>",
"<asp>Campus / Facilities</asp><opn>could not have asked for better</opn><sen>positive</sen>",
"<asp>null</asp><opn>could not have asked for better</opn><sen>positive</sen>",
"<asp>St Germans</asp><opn>get in while you can</opn><sen>positive</sen>",
"<asp>St Germans</asp><opn>nice for us at least</opn><sen>positive</sen>",
"<asp>Gym</asp><opn>usually fine</opn><sen>neutral</sen>",
"<asp>Gym</asp><opn>super fresh</opn><sen>positive</sen>",
"<asp>Gym</asp><opn>can get quite busy</opn><sen>neutral</sen>",
"<asp>actual campus</asp><opn>nice and modern</opn><sen>positive</sen>",
"<asp>Club / societies</asp><opn>makes the year a lot of fun</opn><sen>positive</sen>",
"<asp>Club / societies</asp><opn>smt for everyone</opn><sen>positive</sen>",
"<asp>Club / societies</asp><opn>def get involved</opn><sen>positive</sen>",
"<asp>events</asp><opn>look forward to</opn><sen>positive</sen>",
"<asp>exeter</asp><opn>quite a small town</opn><sen>neutral</sen>"
] |
[
"<asp>first year at Exeter University</asp><opn>absolutely loved it</opn><sen>positive</sen>",
"<asp>Campus / Facilities</asp><opn>could not have asked for better</opn><sen>positive</sen>",
"<asp>null</asp><opn>could not have asked for better</opn><sen>positive</sen>",
"<asp>St Germans</asp><opn>get in while you can</opn><sen>positive</sen>",
"<asp>St Germans</asp><opn>nice for us at least</opn><sen>positive</sen>",
"<asp>Gym</asp><opn>usually fine</opn><sen>neutral</sen>",
"<asp>Gym</asp><opn>super fresh</opn><sen>positive</sen>",
"<asp>Gym</asp><opn>can get quite busy</opn><sen>neutral</sen>",
"<asp>actual campus</asp><opn>nice and modern</opn><sen>positive</sen>",
"<asp>Club / societies</asp><opn>makes the year a lot of fun</opn><sen>positive</sen>",
"<asp>Club / societies</asp><opn>smt for everyone</opn><sen>positive</sen>",
"<asp>Club / societies</asp><opn>def get involved</opn><sen>positive</sen>",
"<asp>events</asp><opn>look forward to</opn><sen>positive</sen>",
"<asp>exeter</asp><opn>quite a small town</opn><sen>neutral</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```I have just completed my first year at Exeter University and I absolutely loved it. Campus / Facilities: I stayed in the St. Germans( sometimes called just Rowe house but they're not the same!) and literally could not have asked for better. I heard that they're going to tear St Germans down to build more blocks like Lafrowda:( i guess its not cost-efficient to have 36 ppl in a block, with 6 ppl per flat with double beds, en-suite, big kitchen+ a nice living room area w bean bags in each flat! But it was nice for us at least so get in while you can. Gym is also new and super fresh, it can get quite busy but there are a lot of different cardio machines and stuff so its usually fine. actual campus is also nice and modern. Club / societies: there is smt for everyone, so def get involved in at least one or two, makes the year a lot of fun and you always have events to look forward to throughout the year, exeter is quite a small town after all so you need to spice your life up a bit!```
|
university review
|
ASTE_1053
|
ASTE
|
1053
|
2_233_54
|
Great course with emphasis on proofs covering many topics. Some of the proofs can be pretty challenging to figure out, but are rewarding upon completion. Start your assignments early and don't be afraid to ask friends or your prof for clues to assignment problems if you are stuck.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>course</asp><opn>Great</opn><sen>positive</sen>",
"<asp>proofs</asp><opn>emphasis</opn><sen>neutral</sen>",
"<asp>topics</asp><opn>many</opn><sen>neutral</sen>",
"<asp>Some of the proofs</asp><opn>pretty challenging</opn><sen>negative</sen>",
"<asp>Some of the proofs</asp><opn>rewarding</opn><sen>positive</sen>"
] |
[
"<asp>course</asp><opn>Great</opn><sen>positive</sen>",
"<asp>proofs</asp><opn>emphasis</opn><sen>neutral</sen>",
"<asp>topics</asp><opn>many</opn><sen>neutral</sen>",
"<asp>Some of the proofs</asp><opn>pretty challenging</opn><sen>negative</sen>",
"<asp>Some of the proofs</asp><opn>rewarding</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Great course with emphasis on proofs covering many topics. Some of the proofs can be pretty challenging to figure out, but are rewarding upon completion. Start your assignments early and don't be afraid to ask friends or your prof for clues to assignment problems if you are stuck.```
|
course review
|
ASTE_1059
|
ASTE
|
1059
|
2_10110_1185
|
Say good bye to your grades for this course! I feel Chong makes this course harder than it should be his slides / annotation were all over the place. I had to more often than not teach myself the course through websites and the text book. That being said despite being taught by Chong I liked the course a lot.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>this course</asp><opn>Say good bye to your grades</opn><sen>negative</sen>",
"<asp>Chong</asp><opn>makes this course harder than it should be</opn><sen>negative</sen>",
"<asp>his slides / annotation</asp><opn>all over the place</opn><sen>negative</sen>",
"<asp>the course</asp><opn>I had to more often than not teach myself</opn><sen>negative</sen>",
"<asp>null</asp><opn>I liked the course a lot</opn><sen>positive</sen>"
] |
[
"<asp>this course</asp><opn>Say good bye to your grades</opn><sen>negative</sen>",
"<asp>Chong</asp><opn>makes this course harder than it should be</opn><sen>negative</sen>",
"<asp>his slides / annotation</asp><opn>all over the place</opn><sen>negative</sen>",
"<asp>the course</asp><opn>I had to more often than not teach myself</opn><sen>negative</sen>",
"<asp>null</asp><opn>I liked the course a lot</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Say good bye to your grades for this course! I feel Chong makes this course harder than it should be his slides / annotation were all over the place. I had to more often than not teach myself the course through websites and the text book. That being said despite being taught by Chong I liked the course a lot.```
|
course review
|
ASTE_1065
|
ASTE
|
1065
|
2_13419_1340
|
Less structure than a lecture, more like a seminar but great!
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>null</asp><opn>Less structure than a lecture</opn><sen>neutral</sen>",
"<asp>null</asp><opn>more like a seminar</opn><sen>neutral</sen>",
"<asp>null</asp><opn>great</opn><sen>positive</sen>"
] |
[
"<asp>null</asp><opn>Less structure than a lecture</opn><sen>neutral</sen>",
"<asp>null</asp><opn>more like a seminar</opn><sen>neutral</sen>",
"<asp>null</asp><opn>great</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Less structure than a lecture, more like a seminar but great!```
|
course review
|
ASTE_1067
|
ASTE
|
1067
|
3_432_329_60
|
I could not imagine going to another University. Exeter has everything, beautiful campus, nice city, near the seaside. Very organized and well run. Spend a lot of efforts helping students find a job when the graduate. Incredible facilities, and one of the best sports complex of any university in the U. K.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>Exeter</asp><opn>could not imagine going to another University</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>has everything</opn><sen>positive</sen>",
"<asp>campus</asp><opn>beautiful</opn><sen>positive</sen>",
"<asp>city</asp><opn>nice</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>Very organized and well run</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>Spend a lot of efforts helping students find a job when the graduate</opn><sen>positive</sen>",
"<asp>facilities</asp><opn>Incredible</opn><sen>positive</sen>",
"<asp>sports complex</asp><opn>one of the best</opn><sen>positive</sen>"
] |
[
"<asp>Exeter</asp><opn>could not imagine going to another University</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>has everything</opn><sen>positive</sen>",
"<asp>campus</asp><opn>beautiful</opn><sen>positive</sen>",
"<asp>city</asp><opn>nice</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>Very organized and well run</opn><sen>positive</sen>",
"<asp>Exeter</asp><opn>Spend a lot of efforts helping students find a job when the graduate</opn><sen>positive</sen>",
"<asp>facilities</asp><opn>Incredible</opn><sen>positive</sen>",
"<asp>sports complex</asp><opn>one of the best</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```I could not imagine going to another University. Exeter has everything, beautiful campus, nice city, near the seaside. Very organized and well run. Spend a lot of efforts helping students find a job when the graduate. Incredible facilities, and one of the best sports complex of any university in the U. K.```
|
university review
|
ASTE_1073
|
ASTE
|
1073
|
2_13711_274
|
Not as hard as people say it is. Overall somewhat complex, but definitely manageable especially if you enjoy the content
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>null</asp><opn>Not as hard as people say it is</opn><sen>positive</sen>",
"<asp>null</asp><opn>Overall somewhat complex</opn><sen>neutral</sen>",
"<asp>null</asp><opn>definitely manageable</opn><sen>positive</sen>"
] |
[
"<asp>null</asp><opn>Not as hard as people say it is</opn><sen>positive</sen>",
"<asp>null</asp><opn>Overall somewhat complex</opn><sen>neutral</sen>",
"<asp>null</asp><opn>definitely manageable</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Not as hard as people say it is. Overall somewhat complex, but definitely manageable especially if you enjoy the content```
|
course review
|
ASTE_1075
|
ASTE
|
1075
|
1_172_2600_11665
|
Professor Mason was the BEST math teacher I have ever had and I am so grateful to have taken her class. I hope that she will teach other Math classes that I need to take. She is really there for her students and you can tell that she likes teaching. Also, a very good teacher if you are a visual learner.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>Professor Mason</asp><opn>a very good teacher</opn><sen>positive</sen>",
"<asp>Professor Mason</asp><opn>the BEST math teacher</opn><sen>positive</sen>",
"<asp>her class</asp><opn>so grateful to have taken</opn><sen>positive</sen>",
"<asp>Professor Mason</asp><opn>hope that she will teach other Math classes</opn><sen>positive</sen>",
"<asp>Professor Mason</asp><opn>really there for her students</opn><sen>positive</sen>",
"<asp>Professor Mason</asp><opn>likes teaching</opn><sen>positive</sen>"
] |
[
"<asp>Professor Mason</asp><opn>a very good teacher</opn><sen>positive</sen>",
"<asp>Professor Mason</asp><opn>the BEST math teacher</opn><sen>positive</sen>",
"<asp>her class</asp><opn>so grateful to have taken</opn><sen>positive</sen>",
"<asp>Professor Mason</asp><opn>hope that she will teach other Math classes</opn><sen>positive</sen>",
"<asp>Professor Mason</asp><opn>really there for her students</opn><sen>positive</sen>",
"<asp>Professor Mason</asp><opn>likes teaching</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Professor Mason was the BEST math teacher I have ever had and I am so grateful to have taken her class. I hope that she will teach other Math classes that I need to take. She is really there for her students and you can tell that she likes teaching. Also, a very good teacher if you are a visual learner.```
|
teacher review
|
ASTE_1079
|
ASTE
|
1079
|
3_421_6_242
|
Absolutely incredible first year, students and staff alike were so helpful and friendly. The amount of societies and clubs on offer means there is something for everyone.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>first year</asp><opn>Absolutely incredible</opn><sen>positive</sen>",
"<asp>students and staff</asp><opn>so helpful and friendly</opn><sen>positive</sen>",
"<asp>societies and clubs</asp><opn>something for everyone</opn><sen>positive</sen>"
] |
[
"<asp>first year</asp><opn>Absolutely incredible</opn><sen>positive</sen>",
"<asp>students and staff</asp><opn>so helpful and friendly</opn><sen>positive</sen>",
"<asp>societies and clubs</asp><opn>something for everyone</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Absolutely incredible first year, students and staff alike were so helpful and friendly. The amount of societies and clubs on offer means there is something for everyone.```
|
university review
|
ASTE_1081
|
ASTE
|
1081
|
3_114_92_283
|
Good social life, nice campus and city. Guild and AI are obstructive and unhelpful
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>social life</asp><opn>Good</opn><sen>positive</sen>",
"<asp>campus and city</asp><opn>nice</opn><sen>positive</sen>",
"<asp>Guild and AI</asp><opn>obstructive and unhelpful</opn><sen>negative</sen>"
] |
[
"<asp>social life</asp><opn>Good</opn><sen>positive</sen>",
"<asp>campus and city</asp><opn>nice</opn><sen>positive</sen>",
"<asp>Guild and AI</asp><opn>obstructive and unhelpful</opn><sen>negative</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Good social life, nice campus and city. Guild and AI are obstructive and unhelpful```
|
university review
|
ASTE_1088
|
ASTE
|
1088
|
2_7869_1301
|
This course was just a nightmare. There is just an overload of information and minute details. But only a few certain ideas and areas are actually tested. I feel like if that is what was expected of us to learn than the prof. should have spent more time on those parts. The material comes hard and fast, with little time for explanation. This course wasn't one I enjoyed, even though I love Molecular Biology. If you can choose not to take this course, I would not take this course.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>This course</asp><opn>nightmare</opn><sen>negative</sen>",
"<asp>information and minute details</asp><opn>overload</opn><sen>negative</sen>",
"<asp>ideas and areas are actually tested</asp><opn>only a few</opn><sen>negative</sen>",
"<asp>prof.</asp><opn>should have spent more time on those parts</opn><sen>negative</sen>",
"<asp>material</asp><opn>little time for explanation</opn><sen>negative</sen>",
"<asp>material</asp><opn>hard and fast</opn><sen>negative</sen>",
"<asp>This course</asp><opn>wasn't one I enjoyed</opn><sen>negative</sen>",
"<asp>Molecular Biology</asp><opn>love</opn><sen>positive</sen>",
"<asp>this course</asp><opn>would not take</opn><sen>negative</sen>"
] |
[
"<asp>This course</asp><opn>nightmare</opn><sen>negative</sen>",
"<asp>information and minute details</asp><opn>overload</opn><sen>negative</sen>",
"<asp>ideas and areas are actually tested</asp><opn>only a few</opn><sen>negative</sen>",
"<asp>prof.</asp><opn>should have spent more time on those parts</opn><sen>negative</sen>",
"<asp>material</asp><opn>little time for explanation</opn><sen>negative</sen>",
"<asp>material</asp><opn>hard and fast</opn><sen>negative</sen>",
"<asp>This course</asp><opn>wasn't one I enjoyed</opn><sen>negative</sen>",
"<asp>Molecular Biology</asp><opn>love</opn><sen>positive</sen>",
"<asp>this course</asp><opn>would not take</opn><sen>negative</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```This course was just a nightmare. There is just an overload of information and minute details. But only a few certain ideas and areas are actually tested. I feel like if that is what was expected of us to learn than the prof. should have spent more time on those parts. The material comes hard and fast, with little time for explanation. This course wasn't one I enjoyed, even though I love Molecular Biology. If you can choose not to take this course, I would not take this course.```
|
course review
|
ASTE_1407
|
ASTE
|
1407
|
3_170_249_271
|
Amazing uni overall with great services
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>uni</asp><opn>Amazing</opn><sen>positive</sen>",
"<asp>services</asp><opn>great</opn><sen>positive</sen>"
] |
[
"<asp>uni</asp><opn>Amazing</opn><sen>positive</sen>",
"<asp>services</asp><opn>great</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Amazing uni overall with great services```
|
university review
|
ASTE_1438
|
ASTE
|
1438
|
3_8_271_251
|
Good, it’s a good university and all the staff are good.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>university</asp><opn>Good</opn><sen>positive</sen>",
"<asp>university</asp><opn>good</opn><sen>positive</sen>",
"<asp>all the staff</asp><opn>good</opn><sen>positive</sen>"
] |
[
"<asp>university</asp><opn>Good</opn><sen>positive</sen>",
"<asp>university</asp><opn>good</opn><sen>positive</sen>",
"<asp>all the staff</asp><opn>good</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Good, it’s a good university and all the staff are good.```
|
university review
|
ASTE_1458
|
ASTE
|
1458
|
3_83_95_47
|
I will miss the time when I stayed here in Exeter. Lovely campus and people.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>Exeter</asp><opn>I will miss the time when I stayed here</opn><sen>positive</sen>",
"<asp>campus</asp><opn>Lovely</opn><sen>positive</sen>",
"<asp>people</asp><opn>Lovely</opn><sen>positive</sen>"
] |
[
"<asp>Exeter</asp><opn>I will miss the time when I stayed here</opn><sen>positive</sen>",
"<asp>campus</asp><opn>Lovely</opn><sen>positive</sen>",
"<asp>people</asp><opn>Lovely</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```I will miss the time when I stayed here in Exeter. Lovely campus and people.```
|
university review
|
ASTE_1483
|
ASTE
|
1483
|
3_513_446_219
|
Most amazing campus, amazing helpful staff and great course content- I absolutely love it here!
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>campus</asp><opn>Most amazing</opn><sen>positive</sen>",
"<asp>staff</asp><opn>amazing</opn><sen>positive</sen>",
"<asp>staff</asp><opn>helpful</opn><sen>positive</sen>",
"<asp>course content</asp><opn>great</opn><sen>positive</sen>",
"<asp>null</asp><opn>absolutely love it here</opn><sen>positive</sen>"
] |
[
"<asp>campus</asp><opn>Most amazing</opn><sen>positive</sen>",
"<asp>staff</asp><opn>amazing</opn><sen>positive</sen>",
"<asp>staff</asp><opn>helpful</opn><sen>positive</sen>",
"<asp>course content</asp><opn>great</opn><sen>positive</sen>",
"<asp>null</asp><opn>absolutely love it here</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Most amazing campus, amazing helpful staff and great course content- I absolutely love it here!```
|
university review
|
ASTE_1487
|
ASTE
|
1487
|
2_6958_294
|
This course had some interesting assignments, such as exploiting a buffer overflow, a format string vulnerability, and finding holes in a poorly secured web app. I found the first half of the course interesting and useful, while the second half felt theoretical and dry. In particular, this course won't teach you how to develop secure web applications if that's what you were looking for.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>assignments</asp><opn>interesting</opn><sen>positive</sen>",
"<asp>first half of the course</asp><opn>interesting and useful</opn><sen>positive</sen>",
"<asp>second half</asp><opn>theoretical and dry</opn><sen>negative</sen>",
"<asp>this course</asp><opn>won't teach you how to develop secure web applications</opn><sen>neutral</sen>"
] |
[
"<asp>assignments</asp><opn>interesting</opn><sen>positive</sen>",
"<asp>first half of the course</asp><opn>interesting and useful</opn><sen>positive</sen>",
"<asp>second half</asp><opn>theoretical and dry</opn><sen>negative</sen>",
"<asp>this course</asp><opn>won't teach you how to develop secure web applications</opn><sen>neutral</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```This course had some interesting assignments, such as exploiting a buffer overflow, a format string vulnerability, and finding holes in a poorly secured web app. I found the first half of the course interesting and useful, while the second half felt theoretical and dry. In particular, this course won't teach you how to develop secure web applications if that's what you were looking for.```
|
course review
|
ASTE_1901
|
ASTE
|
1901
|
2_11277_582
|
I enjoy anatomy and medical studies. This was not an easy course. Studied so hard for all midterms , tests , and exams, and they were absolutely bogus. Quizzed on things I've never even heard of( even after reading each chapter twice and watching lectures twice to understand course content). Written discussion assignments aren't too hard, but final exam made me cry. Especially after spending 21 days studying for 3 hours beforehand.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>anatomy and medical studies</asp><opn>enjoy</opn><sen>positive</sen>",
"<asp>course</asp><opn>not an easy</opn><sen>neutral</sen>",
"<asp>course</asp><opn>Quizzed on things I've never even heard of</opn><sen>negative</sen>",
"<asp>midterms , tests , and exams</asp><opn>absolutely bogus</opn><sen>negative</sen>",
"<asp>Written discussion assignments</asp><opn>aren't too hard</opn><sen>neutral</sen>",
"<asp>final exam</asp><opn>made me cry</opn><sen>negative</sen>"
] |
[
"<asp>anatomy and medical studies</asp><opn>enjoy</opn><sen>positive</sen>",
"<asp>course</asp><opn>not an easy</opn><sen>neutral</sen>",
"<asp>course</asp><opn>Quizzed on things I've never even heard of</opn><sen>negative</sen>",
"<asp>midterms , tests , and exams</asp><opn>absolutely bogus</opn><sen>negative</sen>",
"<asp>Written discussion assignments</asp><opn>aren't too hard</opn><sen>neutral</sen>",
"<asp>final exam</asp><opn>made me cry</opn><sen>negative</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```I enjoy anatomy and medical studies. This was not an easy course. Studied so hard for all midterms , tests , and exams, and they were absolutely bogus. Quizzed on things I've never even heard of( even after reading each chapter twice and watching lectures twice to understand course content). Written discussion assignments aren't too hard, but final exam made me cry. Especially after spending 21 days studying for 3 hours beforehand.```
|
course review
|
ASTE_1905
|
ASTE
|
1905
|
2_8777_460
|
While I generally liked this course, it was really difficult for me to keep up. I'm a fourth year student and even I struggled with the amount of work that was assigned each week. For a 100-level course, the expectations for students are very high and schedule is rigorous. You even lose marks if you miss more than 3 classes, which I found quite unfair especially considering that it was being taught during the pandemic. You will definitely leave with a decent understanding of basic German, but be prepared to do a lot of work. Additionally, the course is taught almost entirely in German from the get-go, and the textbook is written entirely in German. I think this is a huge flaw, as I had to continually translate everything on the internet as we progressed through the term. It's really hard to get a good grasp on the language when everything is introduced and explained in the language you are trying to learn.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>this course</asp><opn>liked</opn><sen>positive</sen>",
"<asp>this course</asp><opn>really difficult</opn><sen>negative</sen>",
"<asp>amount of work</asp><opn>struggled with</opn><sen>negative</sen>",
"<asp>expectations for students</asp><opn>very high</opn><sen>negative</sen>",
"<asp>schedule</asp><opn>rigorous</opn><sen>negative</sen>",
"<asp>lose marks if you miss more than 3 classes</asp><opn>quite unfair</opn><sen>negative</sen>",
"<asp>this course</asp><opn>definitely leave with a decent understanding</opn><sen>positive</sen>",
"<asp>work</asp><opn>be prepared to do a lot</opn><sen>negative</sen>",
"<asp>taught almost entirely in German</asp><opn>huge flaw</opn><sen>negative</sen>",
"<asp>textbook is written entirely in German</asp><opn>huge flaw</opn><sen>negative</sen>",
"<asp>get a good grasp on the language</asp><opn>really hard</opn><sen>negative</sen>"
] |
[
"<asp>this course</asp><opn>liked</opn><sen>positive</sen>",
"<asp>this course</asp><opn>really difficult</opn><sen>negative</sen>",
"<asp>amount of work</asp><opn>struggled with</opn><sen>negative</sen>",
"<asp>expectations for students</asp><opn>very high</opn><sen>negative</sen>",
"<asp>schedule</asp><opn>rigorous</opn><sen>negative</sen>",
"<asp>lose marks if you miss more than 3 classes</asp><opn>quite unfair</opn><sen>negative</sen>",
"<asp>this course</asp><opn>definitely leave with a decent understanding</opn><sen>positive</sen>",
"<asp>work</asp><opn>be prepared to do a lot</opn><sen>negative</sen>",
"<asp>taught almost entirely in German</asp><opn>huge flaw</opn><sen>negative</sen>",
"<asp>textbook is written entirely in German</asp><opn>huge flaw</opn><sen>negative</sen>",
"<asp>get a good grasp on the language</asp><opn>really hard</opn><sen>negative</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```While I generally liked this course, it was really difficult for me to keep up. I'm a fourth year student and even I struggled with the amount of work that was assigned each week. For a 100-level course, the expectations for students are very high and schedule is rigorous. You even lose marks if you miss more than 3 classes, which I found quite unfair especially considering that it was being taught during the pandemic. You will definitely leave with a decent understanding of basic German, but be prepared to do a lot of work. Additionally, the course is taught almost entirely in German from the get-go, and the textbook is written entirely in German. I think this is a huge flaw, as I had to continually translate everything on the internet as we progressed through the term. It's really hard to get a good grasp on the language when everything is introduced and explained in the language you are trying to learn.```
|
course review
|
ASTE_1913
|
ASTE
|
1913
|
1_88_1687_18733
|
All the exam questions come from the assigned homework problems. There are no trick questions. People complained that he doesn't teach. But that is not true! Just read the chapters before class, go to every lecture, and do all the homework. If your still having trouble see him during his office hours. Asking questions in class is too difficult.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>exam questions</asp><opn>no trick questions</opn><sen>positive</sen>",
"<asp>he doesn't teach</asp><opn>not true</opn><sen>positive</sen>",
"<asp>Asking questions in class</asp><opn>too difficult</opn><sen>negative</sen>"
] |
[
"<asp>exam questions</asp><opn>no trick questions</opn><sen>positive</sen>",
"<asp>he doesn't teach</asp><opn>not true</opn><sen>positive</sen>",
"<asp>Asking questions in class</asp><opn>too difficult</opn><sen>negative</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```All the exam questions come from the assigned homework problems. There are no trick questions. People complained that he doesn't teach. But that is not true! Just read the chapters before class, go to every lecture, and do all the homework. If your still having trouble see him during his office hours. Asking questions in class is too difficult.```
|
teacher review
|
ASTE_1918
|
ASTE
|
1918
|
2_9715_1520
|
2 midterms 1 final, all multiple choice. The tests are tricky but the course requires very little effort. It's pretty fun if you take it with your friends because you can gather and watch movies every week. Interesting concepts, deep analysis on film music.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>tests</asp><opn>tricky</opn><sen>negative</sen>",
"<asp>course</asp><opn>requires very little effort</opn><sen>positive</sen>",
"<asp>course</asp><opn>pretty fun</opn><sen>positive</sen>",
"<asp>concepts</asp><opn>Interesting</opn><sen>positive</sen>",
"<asp>analysis on film music</asp><opn>deep</opn><sen>positive</sen>"
] |
[
"<asp>tests</asp><opn>tricky</opn><sen>negative</sen>",
"<asp>course</asp><opn>requires very little effort</opn><sen>positive</sen>",
"<asp>course</asp><opn>pretty fun</opn><sen>positive</sen>",
"<asp>concepts</asp><opn>Interesting</opn><sen>positive</sen>",
"<asp>analysis on film music</asp><opn>deep</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```2 midterms 1 final, all multiple choice. The tests are tricky but the course requires very little effort. It's pretty fun if you take it with your friends because you can gather and watch movies every week. Interesting concepts, deep analysis on film music.```
|
course review
|
ASTE_1920
|
ASTE
|
1920
|
2_9204_320
|
Kwan made this course a nightmare online
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>Kwan</asp><opn>made this course a nightmare</opn><sen>negative</sen>"
] |
[
"<asp>Kwan</asp><opn>made this course a nightmare</opn><sen>negative</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Kwan made this course a nightmare online```
|
course review
|
ASTE_1925
|
ASTE
|
1925
|
2_12141_1498
|
Best course in 1A
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>course</asp><opn>Best</opn><sen>positive</sen>"
] |
[
"<asp>course</asp><opn>Best</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Best course in 1A```
|
course review
|
ASTE_1927
|
ASTE
|
1927
|
1_206_284_13510
|
I love this class. The class discussions are always interesting and helpful. Mrs. Coley really cares about her students, and i learned a lot of life lessons in this class that I will keep with me forever. As long as you study you will pass. You couldnt wish for a better teacher. I would recommend this class to anyone
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>this class</asp><opn>I love</opn><sen>positive</sen>",
"<asp>this class</asp><opn>you will pass</opn><sen>positive</sen>",
"<asp>class discussions</asp><opn>interesting and helpful</opn><sen>positive</sen>",
"<asp>Mrs. Coley</asp><opn>really cares about her students</opn><sen>positive</sen>",
"<asp>this class</asp><opn>i learned a lot of life lessons</opn><sen>positive</sen>",
"<asp>Mrs. Coley</asp><opn>You couldnt wish for a better teacher</opn><sen>positive</sen>",
"<asp>this class</asp><opn>would recommend</opn><sen>positive</sen>"
] |
[
"<asp>this class</asp><opn>I love</opn><sen>positive</sen>",
"<asp>this class</asp><opn>you will pass</opn><sen>positive</sen>",
"<asp>class discussions</asp><opn>interesting and helpful</opn><sen>positive</sen>",
"<asp>Mrs. Coley</asp><opn>really cares about her students</opn><sen>positive</sen>",
"<asp>this class</asp><opn>i learned a lot of life lessons</opn><sen>positive</sen>",
"<asp>Mrs. Coley</asp><opn>You couldnt wish for a better teacher</opn><sen>positive</sen>",
"<asp>this class</asp><opn>would recommend</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```I love this class. The class discussions are always interesting and helpful. Mrs. Coley really cares about her students, and i learned a lot of life lessons in this class that I will keep with me forever. As long as you study you will pass. You couldnt wish for a better teacher. I would recommend this class to anyone```
|
teacher review
|
ASTE_1952
|
ASTE
|
1952
|
2_10808_191
|
A really interesting course with content that you can see come into play during labs and such. Really need to pay attention in lectures and pick up key concepts because exams and lab reports can get pretty nit-picky, but the information isn't too hard to piece together.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>course</asp><opn>really interesting</opn><sen>positive</sen>",
"<asp>content</asp><opn>you can see come into play during labs and such</opn><sen>neutral</sen>",
"<asp>exams and lab reports</asp><opn>can get pretty nit-picky</opn><sen>negative</sen>",
"<asp>information</asp><opn>isn't too hard to piece together</opn><sen>positive</sen>"
] |
[
"<asp>course</asp><opn>really interesting</opn><sen>positive</sen>",
"<asp>content</asp><opn>you can see come into play during labs and such</opn><sen>neutral</sen>",
"<asp>exams and lab reports</asp><opn>can get pretty nit-picky</opn><sen>negative</sen>",
"<asp>information</asp><opn>isn't too hard to piece together</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```A really interesting course with content that you can see come into play during labs and such. Really need to pay attention in lectures and pick up key concepts because exams and lab reports can get pretty nit-picky, but the information isn't too hard to piece together.```
|
course review
|
ASTE_1983
|
ASTE
|
1983
|
3_340_334_156
|
Great experiences at studying here.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>null</asp><opn>Great experiences</opn><sen>positive</sen>"
] |
[
"<asp>null</asp><opn>Great experiences</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Great experiences at studying here.```
|
university review
|
ASTE_1987
|
ASTE
|
1987
|
2_3279_165
|
About half of the course was somewhat review from high school pre-calculus, so the midterm was relatively easy. A lot of the newer content was covered in the second-half of the course but it shouldn't be difficult if you keep up with the assigned work!
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>About half of the course</asp><opn>somewhat review from high school pre-calculus</opn><sen>neutral</sen>",
"<asp>midterm</asp><opn>relatively easy</opn><sen>positive</sen>",
"<asp>newer content</asp><opn>shouldn't be difficult</opn><sen>neutral</sen>"
] |
[
"<asp>About half of the course</asp><opn>somewhat review from high school pre-calculus</opn><sen>neutral</sen>",
"<asp>midterm</asp><opn>relatively easy</opn><sen>positive</sen>",
"<asp>newer content</asp><opn>shouldn't be difficult</opn><sen>neutral</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```About half of the course was somewhat review from high school pre-calculus, so the midterm was relatively easy. A lot of the newer content was covered in the second-half of the course but it shouldn't be difficult if you keep up with the assigned work!```
|
course review
|
ASTE_1993
|
ASTE
|
1993
|
3_24_268_95
|
Great university with very supportive and hardworking lecturers.
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>university</asp><opn>Great</opn><sen>positive</sen>",
"<asp>lecturers</asp><opn>very supportive and hardworking</opn><sen>positive</sen>"
] |
[
"<asp>university</asp><opn>Great</opn><sen>positive</sen>",
"<asp>lecturers</asp><opn>very supportive and hardworking</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Great university with very supportive and hardworking lecturers.```
|
university review
|
ASTE_2201
|
ASTE
|
2201
|
1_172_836_14597
|
awesome math teacher, absolutely hilarious, i loved going to math class. i had the best classmates and he was always making us laugh. he make you feel comfortable and if you don't understand something he explains it thoroughly. i would deff take his class again
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>math teacher</asp><opn>awesome</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>would deff take his class again</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>absolutely hilarious</opn><sen>positive</sen>",
"<asp>math class</asp><opn>i loved going</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>always making us laugh</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>make you feel comfortable</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>explains it thoroughly</opn><sen>positive</sen>"
] |
[
"<asp>math teacher</asp><opn>awesome</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>would deff take his class again</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>absolutely hilarious</opn><sen>positive</sen>",
"<asp>math class</asp><opn>i loved going</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>always making us laugh</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>make you feel comfortable</opn><sen>positive</sen>",
"<asp>math teacher</asp><opn>explains it thoroughly</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```awesome math teacher, absolutely hilarious, i loved going to math class. i had the best classmates and he was always making us laugh. he make you feel comfortable and if you don't understand something he explains it thoroughly. i would deff take his class again```
|
teacher review
|
ASTE_2208
|
ASTE
|
2208
|
1_83_56_11011
|
Very , very informative and overall , great professor. He requires a lot of group work, but assignments are very easy. If you want an easy A, take his course!
|
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
|
[
"<asp>professor</asp><opn>Very , very informative and overall , great</opn><sen>positive</sen>",
"<asp>group work</asp><opn>a lot</opn><sen>neutral</sen>",
"<asp>assignments</asp><opn>very easy</opn><sen>positive</sen>",
"<asp>his course</asp><opn>easy A</opn><sen>positive</sen>"
] |
[
"<asp>professor</asp><opn>Very , very informative and overall , great</opn><sen>positive</sen>",
"<asp>group work</asp><opn>a lot</opn><sen>neutral</sen>",
"<asp>assignments</asp><opn>very easy</opn><sen>positive</sen>",
"<asp>his course</asp><opn>easy A</opn><sen>positive</sen>"
] |
### Task type: aspect-sentiment triplet extraction (ASTE)
### Instruction:
Given the input text, extract ALL pairs of opinion expressions and their corresponding aspect terms about the course, staff, or university. Then classify the sentiment for each aspect-opinion pair.
Opinion expressions are words/phrases expressing evaluation, feeling, or judgment (including both explicit and implicit opinions, not objective facts).
Aspect terms are opinion targets. Only use a pronoun if you cannot find a direct aspect term in the same sentence or adjacent context.
Each aspect-opinion-sentiment combination is a triplet.
**Rules:**
- Extract EVERY opinion in the text, including both explicit and implicit opinion expressions.
- Extract all opinion and aspect terms VERBATIM and as CONSECUTIVE tokens.
- Use 'null' for implicit aspects. Opinions cannot be null.
- If an aspect is mapped to multiple opinion expressions, or vice versa, extract each 1:1 pair separately.
- Classify the sentiment into one of 'positive', 'neutral', 'negative'.
- Use these specific tags for each component within each triplet: <asp>aspect terms</asp>, <opn>opinion expressions</opn>, <sen>sentiment</sen>
**Critical formatting requirements:**
- Output MUST be a valid Python list
- Triplets MUST be separated by commas
**Output format:**
[<asp>...</asp><opn>...</opn><sen>...</sen>, <asp>...</asp><opn>...</opn><sen>...</sen>, ..., <asp>...</asp><opn>...</opn><sen>...</sen>]
### Input:
```Very , very informative and overall , great professor. He requires a lot of group work, but assignments are very easy. If you want an easy A, take his course!```
|
teacher review
|
End of preview.
Licence and Copyright
Copyright (c) 2025 Authors of Data-Efficient Adaptation and a Novel Evaluation Method for Aspect-based Sentiment Analysis.
Both the original, and the formatted versions of the EduRABSA dataset presented in this repository are under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0).
Attribution is required. You may not use this dataset for commercial purposes. Any derivatives must be shared under CC BY-NC-SA 4.0.
EduRABSA_SLM_v1_Test_data
This dataset contains the dev and test sets used for developing the EduRABSA_SLM_v1 models.
This dataset was developed on a subset of the test split of the EduRABSA dataset on the OE, AOPE, AOC, ASTE, and ASQE tasks, where:
- each
devset contains 200 entries per task, and was used for post-training hyperparameter tuning. - each
testset contains 300 entries per task, and was used for final model evaluation. - Each
dev/testdataset has two versions: with 0-shot prompt and 4-shot prompt.
Full details about the dataset are available in the original paper EduRABSA: An Education Review Dataset for Aspect-based Sentiment Analysis Tasks.
To cite this dataset:
@misc{hua2025dataefficientadaptationnovelevaluation,
title={Data-Efficient Adaptation and a Novel Evaluation Method for Aspect-based Sentiment Analysis},
author={Yan Cathy Hua and Paul Denny and Jörg Wicker and Katerina Taškova},
year={2025},
eprint={2511.03034},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2511.03034},
}
- Downloads last month
- 6