{"id":"W3087096149","doi":"10.1016/j.osnem.2020.100096","title":"Hate and offensive speech detection on Arabic social media","year":2020,"lang":"en","type":"article","venue":"Online Social Networks and Media","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":94,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Regina","funders":"","keywords":"Offensive; Computer science; Annotation; Natural language processing; Artificial intelligence; Social media; Class (philosophy); Crawling; Variety (cybernetics); Arabic; Speech recognition; Machine learning; Linguistics; World Wide Web","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":[0.000152158,0.0001763099,0.0002409193,0.00004517285,0.0003883259,0.0001172291,0.0001432731,0.0001926862,0.000006149409],"category_scores_gemma":[0.0001294149,0.0001677098,0.00006082146,0.0003079666,0.00009904106,0.0001329275,0.0001060004,0.0003755268,0.000008117625],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002252964,"about_ca_system_score_gemma":0.00003040442,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001633459,"about_ca_topic_score_gemma":0.0001993287,"domain_scores_codex":[0.9987632,0.00008015746,0.0001801716,0.0004001927,0.0002407459,0.0003355323],"domain_scores_gemma":[0.9993992,0.0001404259,0.00009390424,0.00007488563,0.00007337373,0.0002182459],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006066253,0.00003446617,0.00005021633,0.00001147115,0.0000243486,0.00003322412,0.004260625,0.000023303,0.0004682467,0.001167268,0.001391209,0.992475],"study_design_scores_gemma":[0.01297008,0.003033161,0.1551917,0.0002693509,0.0004361781,0.0002178436,0.007405736,0.6400331,0.007970005,0.01938772,0.1483767,0.004708463],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9324868,0.0005338675,0.04671059,0.015982,0.002796796,0.000372342,0.00003073378,0.0005097895,0.0005770859],"genre_scores_gemma":[0.9920996,0.000330432,0.0009705038,0.001709514,0.004823999,0.000005220146,0.00002483259,0.00001684505,0.00001899389],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9877665,"threshold_uncertainty_score":0.6839009,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02048245754773978,"score_gpt":0.2278576113335182,"score_spread":0.2073751537857784,"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."}}