{"id":"W4389523860","doi":"10.18653/v1/2023.emnlp-main.862","title":"AfriSenti: A Twitter Sentiment Analysis Benchmark for African Languages","year":2023,"lang":"en","type":"article","venue":"","topic":"Linguistics and Language Analysis","field":"Arts and Humanities","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"DeepMind; Universität Hamburg; International Development Research Centre; Rockefeller Foundation","keywords":"Benchmark (surveying); Art; Artificial intelligence; Humanities; Computer science; Art history; Cartography; 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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000197747,0.0001220015,0.0002392818,0.0004492949,0.0002195504,0.0002325733,0.0001271486,0.00002342408,0.01404626],"category_scores_gemma":[0.00004561526,0.00009147011,0.0004041848,0.0003278156,0.00004889453,0.00003442087,0.00004602793,0.0000388588,0.0002304831],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001418644,"about_ca_system_score_gemma":0.00000853751,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009372745,"about_ca_topic_score_gemma":0.002133135,"domain_scores_codex":[0.9990901,0.00001704107,0.0002037918,0.0002375948,0.0001778556,0.000273563],"domain_scores_gemma":[0.9994066,0.000100037,0.00005876986,0.0002483966,0.0001241189,0.0000620725],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00003981751,0.0002433303,0.003912297,0.000109938,0.01033091,0.00008943432,0.1627593,0.0002414217,0.0001684951,0.2621918,0.5565674,0.003345842],"study_design_scores_gemma":[0.0003071341,0.00004536105,0.0005366947,0.000005570444,0.002944198,1.502578e-7,0.05366988,0.007690429,0.0002332059,0.00148643,0.9327269,0.0003540732],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.1940214,0.0002549855,0.001757865,0.002764062,0.0006421286,0.0005518648,0.0003888809,0.000692626,0.7989262],"genre_scores_gemma":[0.8323739,0.000005308936,0.0001624054,0.0003806723,0.000765878,0.00004537435,0.0002732006,0.00001372981,0.1659796],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6383525,"threshold_uncertainty_score":0.986855,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03223315102030863,"score_gpt":0.2847308466247436,"score_spread":0.252497695604435,"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."}}