{"id":"W4311204652","doi":"10.36227/techrxiv.21708188.v1","title":"Regional language toxic comment classification","year":2022,"lang":"en","type":"preprint","venue":"","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Lakehead University","funders":"","keywords":"Marathi; Popularity; Hindi; Social media; Focus (optics); Computer science; Gujarati; Entertainment; Artificial intelligence; Natural language processing; Data science; World Wide Web; Political science; Linguistics","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":[0.0002930065,0.000150151,0.0001317632,0.0001280923,0.0001588439,0.0001842405,0.00104771,0.0001073836,0.0005428914],"category_scores_gemma":[0.00000998424,0.0001485907,0.0001041155,0.0001676728,0.00001585381,0.00009091773,0.001329864,0.0005185943,0.0001183175],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002013302,"about_ca_system_score_gemma":0.0001117974,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001869129,"about_ca_topic_score_gemma":0.0000138181,"domain_scores_codex":[0.9985818,0.0001103921,0.000199553,0.0005181374,0.0004060032,0.0001840686],"domain_scores_gemma":[0.9987553,0.0000325552,0.0001291811,0.0009788083,0.0000331411,0.00007099508],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00002806687,0.0005875337,0.0001910062,0.0001768308,0.0001703908,0.0001170787,0.005531801,0.001557586,0.009060825,0.3692548,0.3276101,0.2857139],"study_design_scores_gemma":[0.0005260599,0.000212173,0.00392611,0.00006281721,0.00002559702,0.00006329394,0.001228512,0.1341243,0.006203684,0.01518179,0.8372931,0.001152522],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01942899,0.0004236752,0.8604267,0.04155672,0.00349075,0.0007601923,0.00001340393,0.001599742,0.07229985],"genre_scores_gemma":[0.8603907,0.000217773,0.1019143,0.01291736,0.0005709467,0.0006401389,0.0004026373,0.00004961445,0.0228966],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8409617,"threshold_uncertainty_score":0.6059354,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03997279582722367,"score_gpt":0.2830121337848864,"score_spread":0.2430393379576627,"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."}}