{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003970552,0.0004963606,0.0002290948,0.002540091,0.0005356616,0.0005763628,0.0001735347,0.0005387011,0.002361456],"category_scores_gemma":[0.001953405,0.00008676913,0.0001998598,0.0006872559,0.0002472623,0.0004328621,0.0005815726,0.0003568512,0.00156996],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002153646,"about_ca_system_score_gemma":0.0002082956,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002124235,"about_ca_topic_score_gemma":0.003321754,"domain_scores_codex":[0.999556,0.0001012272,0.00002461108,0.00006776038,0.0001704877,0.0000797735],"domain_scores_gemma":[0.9985997,0.0005248755,0.0002184873,0.0001089211,0.0003953138,0.0001526777],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00230466,0.0006949639,0.2899293,0.0005922341,0.0001944273,0.003246357,0.002562256,0.00336741,0.1992951,0.001462387,0.01078631,0.4855646],"study_design_scores_gemma":[0.00003150744,0.0008434921,0.6693813,0.0001119366,0.0002064812,0.005040275,0.005534443,0.1928091,0.1095968,0.0010416,0.01526453,0.0001385878],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9821951,0.0004833959,0.008226806,0.000146626,0.0001098243,0.00006560924,0.001150987,0.0004263495,0.007195321],"genre_scores_gemma":[0.9909689,0.0001829192,0.004796636,0.00004049492,0.00007326419,0.00003300683,0.0006524227,0.00001500906,0.003237341],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002540091,"threshold_uncertainty_score":0.007899821,"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."}}