Datasets:

michsethowusu
/

africa-emotions-balanced-multilang

Modalities:

Text

Formats:

Size:

Libraries:

Dataset card Data Studio Files Files and versions

xet

Community

Dataset Viewer

Auto-converted to Parquet

Split (1)

train · 6.04M rows

text stringlengths 1 500	label class label 7 classes
Thou shalt not lie;-D zaidi ya mwaka mmoja uliopita	0sadness
"Ngaqala ukuphila kwami kokusebenza, okokuqala njengomthwalo we-clapper loader, ngakho-ke ngiyidayili yokugxila, bese ngine-opharetha onamafilimu avela lapha eMelbourne.	6disgust
Ga se tshwanelo, monna yo o be a leka go thuša	7neutral
Na izany aza, raha tsikaritrareo fa tara ny fotoana fohy amin'ny dosage manaraka, dia azonao atao ny misintona ny dosage tsy ampy ary mandeha any amin'ny manaraka.	0sadness
Ate Arsenal ya Sanchez, yaakuzannya Man City enkya ku Ssande mu Premier.	6disgust
U kan praat oor hoe u gefrustreerd of kwaad of bang was terwyl die gebeure aan die afspeel was, maar u wil uiteindelik wys dat u met die geleerde les aan die ander kant kon verskyn.	3anger
Wo Wo Wo de Krept and Konan: Napster	5surprise
Nanadihady ny zava-misy manodidina ny mpisera Twitter sasany, ary nanome ny heviny momba izay hahazo ny sezan'ny filoham-pirenena ifampitadiavana fatratra.	5surprise
Qorshaha Dr.Saadiq Eenow...	5surprise
ኢዮአብ እኔን ባሪያህንና የንጉሡን ባሪያ በላከ ጊዜ ትልቅ ሽብር አይቻለሁ፥ ምን እንደሆነ ግን አላወቅሁም ብሎ መለሰለት።	4fear
Ganizirani za zotsatira Ganizirani za patsogolo	3anger
Hali imekuwa mbaya zaidi kwa sisi tunaozipenda timu vigogo.	6disgust
Ku vuriwe leswaku yena i xin'wana xa switirhisiwa swa Xikwembu "emhakeni yo dyondzisa vutivi bya Xikwembu xa ntiyiso na Kreste."	5surprise
Waxaan isku dayaan in ay bixiyaan alaabta LED in shaqayn doonaa in codsi kasta oo hubaal.	7neutral
AungSan cu kawl miphun hrang lawngah a tha	3anger
Khualchhâwn thiamna lantîr nân leh dawn theih nân hun leh tha kan nei tâwk lo a nih chuan eng nge kan tih theih?	6disgust
Toe sy dit die eerste keer vir my, Ek was soos wat dit!	5surprise
Izany no tena be mpampiasa kajy fampiharana sy ny fizarana ny Microsoft Office.	6disgust
W.I.T.C.H. mashabiki (I won't say any names) were insulting Winx mashabiki and I hated it. zaidi ya mwaka mmoja uliopita	3anger
23:3 የእስራኤል አምላክ ተናገረኝ, የእስራኤል ጠንካራ አንዱ ተናገሩ, ሰዎች ገዥ, የ ብቻ ገዥ, በእግዚአብሔር ፍርሃት ውስጥ,	3anger
Lammi gabanaa lammi isaa saaxiluuf safuu.	4fear
11ቃሉ የታመነ ነው እንዲህ የሚለው።	3anger
Lodos waa la xayiray Jimcaha iyo Sabtida sababta oo ah adeegga gawaarida fiilada lama qaban karo maanta.	0sadness
kubela maladi ya sukadi, nitu kukuka diaka ve kutina maladi, yo ke kumisaka maladi mingi, yo ke kumisaka muntu mawa-mawa mpi yo ke kitisaka kilo ya nitu	6disgust
Julle het 'n kompromie aangegaan met satan.	6disgust
Watu waliookolewa katika ajali ya Meli	0sadness
na wɔrentumi mfa deɛ wopɛ biara ntoto ho.	1joy
ዘካርያስ በኢየሩሳሌም በሚገኘው የአምላክ ቤተ መቅደስ ውስጥ በክህነት የሚያገለግልበት ተራ ደርሷል።	4fear
jis jeevan mein aa gale lag jaa full song	4fear
Aliwesi abaghiye ukusala mwene ukwivwila ilamulo ilya Chala pa nkani iya ibanda.	0sadness
Nee Ukku Laanti Onti Theeru Greeky Shilpame	1joy
እሷ በጣም ሐቀኛ ስለሆነች ሁል ጊዜ ወደ አማንዳ እመለሳለሁ ።	1joy
Adura ti eeyan maa nse ti eyan ba ni ipaya kan ni oju orun abi eniti won bafi adanwo ipa aayun se	3anger
"Ryan en sy familie het hul waardering uitgespreek vir die boodskappe van beterskap en mense se besorgdheid van oor die hele land heen.	4fear
Casanis a yafe awalc'h ganin.	7neutral
"Haba the juiciest one ina zaki....zo ki fadamin how am I in bed..."Yafada yana wullata kan gado	0sadness
Cindy Crawford sere serpe - Magazin - Mahmure Foto Galeri	7neutral
Au mlikuwa muntiya asumini na	7neutral
Mbandu ni jingivuidi, ia lungu ni "Ifika ia Bhebhuluka."	3anger
ooo je veux vivrree en malaiisiie xD	4fear
Na, licha ya mengine mengi, kila siku nakabiliwa na shughuli za makanisa yote.	4fear
"Ita kenan ba Dan abinda Tai min Kai Mata haka ba ko?"	4fear
Hiki ni kilio kikubwa kwa wananchi wanaopisha ujenzi wa mradi huu.	6disgust
voordat dit was getrek uit die agenda in maart die ahca onder ander dinge gesoek te vervang obamacare s netwerk van versekering premium subsidies met 'n beduidende expansion in die gebruik van belasting krediete en gesondheid spaar rekeninge	7neutral
Shy, mbona hujibu?	3anger
Ova pewa "omhepo" nova teelela 'okutambulwa moudalwa womalutu avo opambelela.'	6disgust
Hoe't jy geweet, hmm?	5surprise
kid buu eza weakness	4fear
Maye Amento	6disgust
Hello! afeekkk interesting afeekkk site!	1joy
kya tum mudogi kabhi,	4fear
Saropady ny volamena amintsika Malagasy, ka tsy azo vazivaziana	7neutral
Ary na inona na inona no hitranga dia tsy hisy intsony ny "premier tour dia vita."	0sadness
enwiki Lõpemetsa	3anger
lala leche lala jugo lala café	6disgust
Newcastle Go Karts Go	5surprise
Cavità Interne o Esterne	1joy
Hujui nini kinachozungumziwa.	5surprise
ary ny zava-misy iainanao.	0sadness
Bango bien dix sur dix bango ba koka na tout kasi masolo nyoso ya mokili kaka se ya Mopao ehh toujours	7neutral
vhatshinyi a vha nga dzheni khuvhanganoni ya vhavhuya.	6disgust
Tianay ny hahita ny indostria misy antsika mamirapiratra toy ny manga, maitso ary fotsy izay maneho ny hatsaran'ny firenena.	0sadness
Iyo deriska kala yaacayeen; dib isu raadcaynin	6disgust
Ummeli Wokuhlwaya (iSpanish)	5surprise
Sihlale le eParadisi,	1joy
N'cyetu benë be n'kwambililë: " Eeh, ya cisilikilë ci libakëlë, n'tininë kwa singasi osi e! " (minaanu matë)	1joy
ገንዘቡ ቀበሩን ለማስፈጸም እያቃተው ነው ።	3anger
Ek was verbaas om oplossing te vind vir al die probleme in een instrument.	5surprise
Side ayad ado habeno dhan so jeeda wax u dhigan karta Malinti?	5surprise
Na wale wanene walio kwenye ndoa wameingiaje sasa?	7neutral
"Aaaiiii, Aaaiiii" seklinde yapiyordu ki, beni en cok	4fear
aaf unlogged	4fear
Innteakte puak khat po na nei le pia paih in.	6disgust
Ananse ufunile inhlakanipho ka umhlaba.	0sadness
Ko e nder maggal Duunde Jeeriyel ndee woni.	6disgust
Mimi najua iliko asali iliyo nzuri.	7neutral
Ny harena, hita avy any ambony	5surprise
Dil ko haay dil ko (Daastaan)	0sadness
Ary any amin'ireo nofy lavitra	4fear
Lawyer Wanza Kivindu	0sadness
ከጫጩቱ ውስጥ የሚወጣውን ፈሳሽ አፍጥጠው ይጥሉት, እንደከፊቱ ያደርጉታል.	7neutral
moku: "pipi mute o! mi wile moku!"	3anger
Na Salamu Allah Juu Ya Wote Ma Anssar Walio Tangulia Walio Bora Na Juu Ya Waislamu Wote, Hakika Mola Mlezi wangu Ni Mpana Wa Huruma Na Maghfira, Hakika Mola Mlezi Wangu Ni Ghafur Mrahimu..	1joy
tzhost inakupa yote hayo pamoja na haya yafuatayo....	7neutral
በዚህ ዓመት የተዘራው አከባቢ 200 ሺህ ሄክታር ነው ።	0sadness
Hadaba sida loo isticmaalo shabakadan cusub hoos kadaawo. Mahadsanid	7neutral
Jana tumak paboloi	4fear
ለአራት ቀን ከተያዙት ልጆች ጋር መገናኘት አልቻልኩም።	4fear
n'ai-je pas aboyé ?	5surprise
V መረጃ ነው .	7neutral
Nadi, layizgo lakuti caru cizamuŵa paradiso likanovwira kuti nisinthe umoyo wane.	3anger
Gara fuula duraa tarkaanachuu dhiisanii bitaa of irraa galagalanii gara boodaa deemaa jiru.	0sadness
A Bit About Emmy	7neutral
እስከዚያው ግን ከፋይ የሚዘራው ያጣው ገበሬ ነው።	4fear
Afisin'ny atrikasa.	7neutral
Leo's is die tipe mense wat hul hare in 'n mal nuwe styl kan dra (net omdat hulle wil!), En binne weke het dit die nuwe neiging geword.	5surprise
ኣንድ single የአምራቾች የህብረት ሥራ ማህበር አለ ወይ ትግሬ ውስጥ?	1joy
Look: Magelo Black	7neutral
የፒሪ ሬይስ ካርታ በአሁኑ ጊዜ በቱርክ ኢስታንቡል ውስጥ ባለው Topkapi ቤተመፃሕፍት ውስጥ ይገኛል ፣ ግን በአሁኑ ጊዜ በእይታ ላይ አይደለም።	7neutral
mosansimobi kukusxs	7neutral

End of preview. Expand in Data Studio

Multilingual Emotion Analysis Corpus

Dataset Description

This dataset contains emotion-labeled text data in multiple African languages for emotion classification (joy, sadness, anger, fear, surprise, disgust, neutral). Emotions were inferred from the original content using a multilingual processing pipeline and balanced per language to support fairer modeling.

The dataset is part of a growing collection of African language emotion analysis resources.

Dataset Statistics

Language: afrikaans Total samples: 511287

Anger: 73041 (14.3%)
Disgust: 73041 (14.3%)
Fear: 73041 (14.3%)
Joy: 73041 (14.3%)
Neutral: 73041 (14.3%)
Sadness: 73041 (14.3%)
Surprise: 73041 (14.3%) Language: akan Total samples: 1694
Anger: 242 (14.3%)
Disgust: 242 (14.3%)
Fear: 242 (14.3%)
Joy: 242 (14.3%)
Neutral: 242 (14.3%)
Sadness: 242 (14.3%)
Surprise: 242 (14.3%) Language: amharic Total samples: 568750
Anger: 81250 (14.3%)
Disgust: 81250 (14.3%)
Fear: 81250 (14.3%)
Joy: 81250 (14.3%)
Neutral: 81250 (14.3%)
Sadness: 81250 (14.3%)
Surprise: 81250 (14.3%) Language: bambara Total samples: 15309
Anger: 2187 (14.3%)
Disgust: 2187 (14.3%)
Fear: 2187 (14.3%)
Joy: 2187 (14.3%)
Neutral: 2187 (14.3%)
Sadness: 2187 (14.3%)
Surprise: 2187 (14.3%) Language: bemba Total samples: 47985
Anger: 6855 (14.3%)
Disgust: 6855 (14.3%)
Fear: 6855 (14.3%)
Joy: 6855 (14.3%)
Neutral: 6855 (14.3%)
Sadness: 6855 (14.3%)
Surprise: 6855 (14.3%) Language: chichewa Total samples: 222166
Anger: 31738 (14.3%)
Disgust: 31738 (14.3%)
Fear: 31738 (14.3%)
Joy: 31738 (14.3%)
Neutral: 31738 (14.3%)
Sadness: 31738 (14.3%)
Surprise: 31738 (14.3%) Language: dinka Total samples: 9611
Anger: 1373 (14.3%)
Disgust: 1373 (14.3%)
Fear: 1373 (14.3%)
Joy: 1373 (14.3%)
Neutral: 1373 (14.3%)
Sadness: 1373 (14.3%)
Surprise: 1373 (14.3%) Language: dyula Total samples: 13874
Anger: 1982 (14.3%)
Disgust: 1982 (14.3%)
Fear: 1982 (14.3%)
Joy: 1982 (14.3%)
Neutral: 1982 (14.3%)
Sadness: 1982 (14.3%)
Surprise: 1982 (14.3%) Language: ewe Total samples: 116326
Anger: 16618 (14.3%)
Disgust: 16618 (14.3%)
Fear: 16618 (14.3%)
Joy: 16618 (14.3%)
Neutral: 16618 (14.3%)
Sadness: 16618 (14.3%)
Surprise: 16618 (14.3%) Language: fon Total samples: 14035
Anger: 2005 (14.3%)
Disgust: 2005 (14.3%)
Fear: 2005 (14.3%)
Joy: 2005 (14.3%)
Neutral: 2005 (14.3%)
Sadness: 2005 (14.3%)
Surprise: 2005 (14.3%) Language: fulah Total samples: 17955
Anger: 2565 (14.3%)
Disgust: 2565 (14.3%)
Fear: 2565 (14.3%)
Joy: 2565 (14.3%)
Neutral: 2565 (14.3%)
Sadness: 2565 (14.3%)
Surprise: 2565 (14.3%) Language: ganda Total samples: 227689
Anger: 32527 (14.3%)
Disgust: 32527 (14.3%)
Fear: 32527 (14.3%)
Joy: 32527 (14.3%)
Neutral: 32527 (14.3%)
Sadness: 32527 (14.3%)
Surprise: 32527 (14.3%) Language: hausa Total samples: 207732
Anger: 29676 (14.3%)
Disgust: 29676 (14.3%)
Fear: 29676 (14.3%)
Joy: 29676 (14.3%)
Neutral: 29676 (14.3%)
Sadness: 29676 (14.3%)
Surprise: 29676 (14.3%) Language: igbo Total samples: 49350
Anger: 7050 (14.3%)
Disgust: 7050 (14.3%)
Fear: 7050 (14.3%)
Joy: 7050 (14.3%)
Neutral: 7050 (14.3%)
Sadness: 7050 (14.3%)
Surprise: 7050 (14.3%) Language: kabiye Total samples: 5411
Anger: 773 (14.3%)
Disgust: 773 (14.3%)
Fear: 773 (14.3%)
Joy: 773 (14.3%)
Neutral: 773 (14.3%)
Sadness: 773 (14.3%)
Surprise: 773 (14.3%) Language: kabuverdianu Total samples: 30072
Anger: 4296 (14.3%)
Disgust: 4296 (14.3%)
Fear: 4296 (14.3%)
Joy: 4296 (14.3%)
Neutral: 4296 (14.3%)
Sadness: 4296 (14.3%)
Surprise: 4296 (14.3%) Language: kabyle Total samples: 1575
Anger: 225 (14.3%)
Disgust: 225 (14.3%)
Fear: 225 (14.3%)
Joy: 225 (14.3%)
Neutral: 225 (14.3%)
Sadness: 225 (14.3%)
Surprise: 225 (14.3%) Language: kamba Total samples: 8407
Anger: 1201 (14.3%)
Disgust: 1201 (14.3%)
Fear: 1201 (14.3%)
Joy: 1201 (14.3%)
Neutral: 1201 (14.3%)
Sadness: 1201 (14.3%)
Surprise: 1201 (14.3%) Language: kikuyu Total samples: 13160
Anger: 1880 (14.3%)
Disgust: 1880 (14.3%)
Fear: 1880 (14.3%)
Joy: 1880 (14.3%)
Neutral: 1880 (14.3%)
Sadness: 1880 (14.3%)
Surprise: 1880 (14.3%) Language: kimbundu Total samples: 20034
Anger: 2862 (14.3%)
Disgust: 2862 (14.3%)
Fear: 2862 (14.3%)
Joy: 2862 (14.3%)
Neutral: 2862 (14.3%)
Sadness: 2862 (14.3%)
Surprise: 2862 (14.3%) Language: kongo Total samples: 47894
Anger: 6842 (14.3%)
Disgust: 6842 (14.3%)
Fear: 6842 (14.3%)
Joy: 6842 (14.3%)
Neutral: 6842 (14.3%)
Sadness: 6842 (14.3%)
Surprise: 6842 (14.3%) Language: lingala Total samples: 132671
Anger: 18953 (14.3%)
Disgust: 18953 (14.3%)
Fear: 18953 (14.3%)
Joy: 18953 (14.3%)
Neutral: 18953 (14.3%)
Sadness: 18953 (14.3%)
Surprise: 18953 (14.3%) Language: lushai Total samples: 152495
Anger: 21785 (14.3%)
Disgust: 21785 (14.3%)
Fear: 21785 (14.3%)
Joy: 21785 (14.3%)
Neutral: 21785 (14.3%)
Sadness: 21785 (14.3%)
Surprise: 21785 (14.3%) Language: malagasy Total samples: 388367
Anger: 55481 (14.3%)
Disgust: 55481 (14.3%)
Fear: 55481 (14.3%)
Joy: 55481 (14.3%)
Neutral: 55481 (14.3%)
Sadness: 55481 (14.3%)
Surprise: 55481 (14.3%) Language: mossi Total samples: 38787
Anger: 5541 (14.3%)
Disgust: 5541 (14.3%)
Fear: 5541 (14.3%)
Joy: 5541 (14.3%)
Neutral: 5541 (14.3%)
Sadness: 5541 (14.3%)
Surprise: 5541 (14.3%) Language: nuer Total samples: 2415
Anger: 345 (14.3%)
Disgust: 345 (14.3%)
Fear: 345 (14.3%)
Joy: 345 (14.3%)
Neutral: 345 (14.3%)
Sadness: 345 (14.3%)
Surprise: 345 (14.3%) Language: oromo Total samples: 242879
Anger: 34697 (14.3%)
Disgust: 34697 (14.3%)
Fear: 34697 (14.3%)
Joy: 34697 (14.3%)
Neutral: 34697 (14.3%)
Sadness: 34697 (14.3%)
Surprise: 34697 (14.3%) Language: pedi Total samples: 115255
Anger: 16465 (14.3%)
Disgust: 16465 (14.3%)
Fear: 16465 (14.3%)
Joy: 16465 (14.3%)
Neutral: 16465 (14.3%)
Sadness: 16465 (14.3%)
Surprise: 16465 (14.3%) Language: rundi Total samples: 125419
Anger: 17917 (14.3%)
Disgust: 17917 (14.3%)
Fear: 17917 (14.3%)
Joy: 17917 (14.3%)
Neutral: 17917 (14.3%)
Sadness: 17917 (14.3%)
Surprise: 17917 (14.3%) Language: sango Total samples: 33467
Anger: 4781 (14.3%)
Disgust: 4781 (14.3%)
Fear: 4781 (14.3%)
Joy: 4781 (14.3%)
Neutral: 4781 (14.3%)
Sadness: 4781 (14.3%)
Surprise: 4781 (14.3%) Language: somali Total samples: 670327
Anger: 95761 (14.3%)
Disgust: 95761 (14.3%)
Fear: 95761 (14.3%)
Joy: 95761 (14.3%)
Neutral: 95761 (14.3%)
Sadness: 95761 (14.3%)
Surprise: 95761 (14.3%) Language: swahili Total samples: 626913
Anger: 89559 (14.3%)
Disgust: 89559 (14.3%)
Fear: 89559 (14.3%)
Joy: 89559 (14.3%)
Neutral: 89559 (14.3%)
Sadness: 89559 (14.3%)
Surprise: 89559 (14.3%) Language: swati Total samples: 30198
Anger: 4314 (14.3%)
Disgust: 4314 (14.3%)
Fear: 4314 (14.3%)
Joy: 4314 (14.3%)
Neutral: 4314 (14.3%)
Sadness: 4314 (14.3%)
Surprise: 4314 (14.3%) Language: tamasheq Total samples: 119
Anger: 17 (14.3%)
Disgust: 17 (14.3%)
Fear: 17 (14.3%)
Joy: 17 (14.3%)
Neutral: 17 (14.3%)
Sadness: 17 (14.3%)
Surprise: 17 (14.3%) Language: tsonga Total samples: 84728
Anger: 12104 (14.3%)
Disgust: 12104 (14.3%)
Fear: 12104 (14.3%)
Joy: 12104 (14.3%)
Neutral: 12104 (14.3%)
Sadness: 12104 (14.3%)
Surprise: 12104 (14.3%) Language: tswana Total samples: 271264
Anger: 38752 (14.3%)
Disgust: 38752 (14.3%)
Fear: 38752 (14.3%)
Joy: 38752 (14.3%)
Neutral: 38752 (14.3%)
Sadness: 38752 (14.3%)
Surprise: 38752 (14.3%) Language: tumbuka Total samples: 64869
Anger: 9267 (14.3%)
Disgust: 9267 (14.3%)
Fear: 9267 (14.3%)
Joy: 9267 (14.3%)
Neutral: 9267 (14.3%)
Sadness: 9267 (14.3%)
Surprise: 9267 (14.3%) Language: twi Total samples: 140917
Anger: 20131 (14.3%)
Disgust: 20131 (14.3%)
Fear: 20131 (14.3%)
Joy: 20131 (14.3%)
Neutral: 20131 (14.3%)
Sadness: 20131 (14.3%)
Surprise: 20131 (14.3%) Language: umbundu Total samples: 27608
Anger: 3944 (14.3%)
Disgust: 3944 (14.3%)
Fear: 3944 (14.3%)
Joy: 3944 (14.3%)
Neutral: 3944 (14.3%)
Sadness: 3944 (14.3%)
Surprise: 3944 (14.3%) Language: wolof Total samples: 92820
Anger: 13260 (14.3%)
Disgust: 13260 (14.3%)
Fear: 13260 (14.3%)
Joy: 13260 (14.3%)
Neutral: 13260 (14.3%)
Sadness: 13260 (14.3%)
Surprise: 13260 (14.3%) Language: xhosa Total samples: 536837
Anger: 76691 (14.3%)
Disgust: 76691 (14.3%)
Fear: 76691 (14.3%)
Joy: 76691 (14.3%)
Neutral: 76691 (14.3%)
Sadness: 76691 (14.3%)
Surprise: 76691 (14.3%) Language: yoruba Total samples: 55223
Anger: 7889 (14.3%)
Disgust: 7889 (14.3%)
Fear: 7889 (14.3%)
Joy: 7889 (14.3%)
Neutral: 7889 (14.3%)
Sadness: 7889 (14.3%)
Surprise: 7889 (14.3%) Language: zulu Total samples: 59990
Anger: 8570 (14.3%)
Disgust: 8570 (14.3%)
Fear: 8570 (14.3%)
Joy: 8570 (14.3%)
Neutral: 8570 (14.3%)
Sadness: 8570 (14.3%)
Surprise: 8570 (14.3%)

Dataset Structure

Data Fields

text: Original text
emotion: Emotion label (joy, sadness, anger, fear, surprise, disgust, neutral)
language: Language code (e.g., twi, am, sw, etc.)

Data Splits

This dataset contains a single split with all the balanced data.

Data Processing

Label Generation: Used existing datasets and automatic labeling pipelines.
Balancing: Emotion classes were trimmed to match the smallest class per language.
Deduplication: Removed duplicate texts.
Language Coverage: Includes multiple African languages, including those using non-Latin scripts.

Usage

from datasets import load_dataset
dataset = load_dataset("michsethowusu/africa-emotions-balanced-multilang")
print(dataset['train'][0])

Citation

If you use this dataset in your research, please cite:

@dataset{emotion_multilang_africa,
  title={Multilingual African Emotion Corpus},
  author={Your Name},
  year={2025},
  url={https://huggingface.co/datasets/michsethowusu/africa-emotions-balanced-multilang}
}