{"id":"W2955549617","doi":"10.18653/v1/w18-5103","title":"A Review of Standard Text Classification Practices for Multi-label Toxicity Identification of Online Content","year":2018,"lang":"en","type":"review","venue":"","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":38,"is_retracted":false,"has_abstract":true,"ca_institutions":"International Medias Data Services (Canada)","funders":"","keywords":"Computer science; Artificial intelligence; Classifier (UML); Support vector machine; Inference; Natural language processing; Social media; Machine learning; Language identification; Speech recognition; Natural language; 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":[0.001856204,0.0002574409,0.001231754,0.0001803177,0.00006531036,0.00005332051,0.0008924272,0.0001959881,0.00001295863],"category_scores_gemma":[0.001728706,0.0001953573,0.0003670431,0.0005786292,0.00007476864,0.0003976942,0.0001031825,0.0001435178,0.00001913469],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001073928,"about_ca_system_score_gemma":0.0004436826,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003093082,"about_ca_topic_score_gemma":0.00003223424,"domain_scores_codex":[0.9969661,0.00024855,0.001680385,0.0005673972,0.0003592882,0.0001782354],"domain_scores_gemma":[0.9919213,0.0002423609,0.005472234,0.0009703781,0.001330893,0.00006285301],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000005359834,0.0002014285,1.529716e-7,0.08892681,0.00006708112,9.111584e-8,0.00001204632,8.136643e-9,0.0001504457,0.001163387,0.001000879,0.9084723],"study_design_scores_gemma":[0.0002738891,0.0003292028,0.000003142199,0.05853752,0.0004402497,0.00001058691,0.00001387837,0.00319147,0.0009159379,0.00003003111,0.9360151,0.0002389379],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.000001226843,0.6494019,0.348572,0.00009250313,0.0002470982,0.001472119,0.0001223228,0.00004450658,0.00004626379],"genre_scores_gemma":[0.000001519795,0.8875304,0.1114455,0.00006181873,0.00006247811,0.0001492621,0.000118889,0.0000158498,0.0006142521],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9350142,"threshold_uncertainty_score":0.7966437,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3275048838604004,"score_gpt":0.4401816853434807,"score_spread":0.1126768014830802,"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."}}