{"id":"W3209074625","doi":"10.1080/08839514.2021.1988443","title":"Semi-Supervised Self-Training of Hate and Offensive Speech from Social Media","year":2021,"lang":"en","type":"article","venue":"Applied Artificial Intelligence","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Regina","funders":"","keywords":"Computer science; Classifier (UML); Artificial intelligence; Offensive; Leverage (statistics); Social media; Machine learning; Support vector machine; Arabic; Training set; Supervised learning; Labeled data; Semi-supervised learning; Natural language processing; Speech recognition; Pattern recognition (psychology); 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":[0.002374754,0.001211938,0.0009404859,0.001029108,0.0006407858,0.000804153,0.001200205,0.001050228,0.001379895],"category_scores_gemma":[0.005670395,0.0002773828,0.0006673075,0.0003953722,0.0006087312,0.001532697,0.001146954,0.00152869,0.001883787],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004635087,"about_ca_system_score_gemma":0.0007691383,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001948482,"about_ca_topic_score_gemma":0.004327036,"domain_scores_codex":[0.9986703,0.0004569845,0.00008465344,0.000428649,0.0002121389,0.0001472083],"domain_scores_gemma":[0.9946695,0.002450937,0.0004538132,0.0008007819,0.001372146,0.0002528222],"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":[0.001008831,0.002390378,0.04556883,0.000365985,0.000277099,0.000214924,0.0006790173,0.0679713,0.0420326,0.001091397,0.01358026,0.8248194],"study_design_scores_gemma":[0.00002598811,0.0003447756,0.01060827,0.00003151502,0.00004039226,0.0001233097,0.0002729218,0.9498544,0.03563095,0.001137668,0.001893297,0.00003649234],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7592961,0.0005947878,0.2235089,0.0004719724,0.0003188443,0.000480033,0.001276073,0.006179556,0.00787375],"genre_scores_gemma":[0.9229438,0.00008365404,0.06908265,0.000167995,0.00008136199,0.0002007449,0.00304231,0.0001186973,0.004278796],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002374754,"threshold_uncertainty_score":0.01255906,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04537766661798438,"score_gpt":0.2479983857249821,"score_spread":0.2026207191069977,"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."}}