{"id":"W4321392937","doi":"10.48550/arxiv.2302.08956","title":"AfriSenti: A Twitter Sentiment Analysis Benchmark for African Languages","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Sentiment Analysis and Opinion Mining","field":"Computer Science","cited_by":47,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"DeepMind; Universität Hamburg; International Development Research Centre; Rockefeller Foundation","keywords":"Amharic; Computer science; Arabic languages; Natural language processing; Task (project management); Languages of Africa; Artificial intelligence; Portuguese; Annotation; Benchmark (surveying); Linguistics; Arabic; World Wide Web; Geography","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004544031,0.0003701503,0.0006217507,0.001199838,0.0002042408,0.0003247829,0.001694786,0.0002075099,0.0001234037],"category_scores_gemma":[0.00002418607,0.0004043732,0.001169966,0.002426779,0.00005574025,0.0002066524,0.001841951,0.0002661189,0.0001173173],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001514821,"about_ca_system_score_gemma":0.00007928133,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000193188,"about_ca_topic_score_gemma":0.00007519864,"domain_scores_codex":[0.9973423,0.0001201609,0.0003257709,0.00153277,0.0001857297,0.0004933072],"domain_scores_gemma":[0.9975999,0.0001976493,0.0003964451,0.001455988,0.000169337,0.0001806834],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001078733,0.0006780915,0.09818625,0.000372195,0.02374477,0.0007030015,0.005799972,0.7284206,0.0001892886,0.09544526,0.04459492,0.00175781],"study_design_scores_gemma":[0.0005243572,0.00004180482,0.00566111,0.00005591797,0.002490626,4.303217e-7,0.0009734596,0.9798285,0.0002126177,0.006887335,0.002568091,0.000755779],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1062246,0.00008169234,0.8903049,0.000545004,0.0006164853,0.0004281995,0.00004588656,0.0003692209,0.001383996],"genre_scores_gemma":[0.9803079,0.00006235217,0.004569631,0.00009844006,0.0001435116,0.000006746935,0.0001838651,0.00002452109,0.01460308],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8857352,"threshold_uncertainty_score":0.9998408,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1100176813714993,"score_gpt":0.2375084513938683,"score_spread":0.127490770022369,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}