{"id":"W4392669860","doi":"10.18653/v1/2023.ijcnlp-main.10","title":"MasakhaNEWS: News Topic Classification for African languages","year":2023,"lang":"en","type":"article","venue":"","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute; University of Waterloo","funders":"DeepMind","keywords":"Computer science; Languages of Africa; Natural language processing; Artificial intelligence; Linguistics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001127327,0.00007014954,0.00007505851,0.0001648611,0.00008056786,0.0001246917,0.0006395038,0.00005179256,0.00002302684],"category_scores_gemma":[0.00008340969,0.00005793105,0.000042026,0.0005895771,0.00002461351,0.0002525262,0.0001043739,0.00003660432,0.0002803051],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002245034,"about_ca_system_score_gemma":0.00002026265,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001575813,"about_ca_topic_score_gemma":0.00001683457,"domain_scores_codex":[0.9992981,0.0000111661,0.0001350234,0.000259227,0.0001117499,0.0001847178],"domain_scores_gemma":[0.9992732,0.0000980737,0.00005140729,0.0005113359,0.00003660471,0.00002933935],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[9.901377e-7,0.00001078927,0.0004190927,0.00000650312,0.00000381913,5.520779e-7,0.0001564886,0.000001548332,0.002609421,0.7358885,0.04046857,0.2204337],"study_design_scores_gemma":[0.0006408709,0.0001627974,0.0366671,0.000009683935,0.000008042339,0.000002703669,0.007380658,0.04852069,0.02345815,0.109434,0.7732463,0.0004690138],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005438406,0.00005071828,0.9091445,0.04733341,0.0002633187,0.0004172356,0.000003081622,0.004705747,0.03264362],"genre_scores_gemma":[0.9313816,0.00004866131,0.04012151,0.0003271866,0.00004698272,0.000244924,0.00001863563,0.000007589083,0.02780293],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9259432,"threshold_uncertainty_score":0.3602848,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0562600828557314,"score_gpt":0.3205969545643245,"score_spread":0.264336871708593,"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."}}