{"id":"W4206166506","doi":"10.1109/smc52423.2021.9659134","title":"Semi-Supervised Self-Learning for Arabic Hate Speech Detection","year":2021,"lang":"en","type":"article","venue":"2021 IEEE International Conference on Systems, Man, and Cybernetics (SMC)","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Regina","funders":"","keywords":"Computer science; Arabic; Classifier (UML); Voice activity detection; Artificial intelligence; Speech recognition; Supervised learning; Natural language processing; Social media; Training set; Machine learning; Speech processing; Linguistics; World Wide Web; Artificial neural network","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":[0.002792614,0.001242124,0.001209916,0.001259376,0.0008838262,0.0008982508,0.001769077,0.001210943,0.002187346],"category_scores_gemma":[0.008185772,0.0003867331,0.0007247623,0.0006609348,0.0009962223,0.001726001,0.001492208,0.001906384,0.003357193],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005181439,"about_ca_system_score_gemma":0.000802157,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001554422,"about_ca_topic_score_gemma":0.003696553,"domain_scores_codex":[0.9977738,0.0008810786,0.0001427144,0.0006669455,0.0003832858,0.0001521768],"domain_scores_gemma":[0.990682,0.0045884,0.0007357771,0.001386969,0.002359435,0.0002473552],"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":[0.0007778465,0.001157105,0.01029513,0.0003651255,0.0001919783,0.000257871,0.0005820925,0.05354349,0.04555986,0.001903266,0.01429629,0.8710701],"study_design_scores_gemma":[0.00002678436,0.0001944895,0.00297094,0.00002384575,0.00003001417,0.0001663281,0.0001684035,0.9606679,0.03001489,0.002765167,0.002928918,0.00004232167],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1687534,0.0006782772,0.8149111,0.0004035235,0.0002341766,0.0004019034,0.0008841384,0.008476582,0.005256947],"genre_scores_gemma":[0.7528174,0.0001788104,0.2345103,0.0003767053,0.0002018482,0.0004847215,0.003812199,0.0003500817,0.007268005],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002792614,"threshold_uncertainty_score":0.0147689,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03250911837304178,"score_gpt":0.2649787246159093,"score_spread":0.2324696062428675,"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."}}