{"id":"W4205675155","doi":"10.1109/bigdata52589.2021.9672023","title":"MetaHate: A Meta-Model for Hate Speech Detection","year":2021,"lang":"en","type":"article","venue":"2021 IEEE International Conference on Big Data (Big Data)","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Carleton University","funders":"","keywords":"Benchmark (surveying); Computer science; Baseline (sea); Task (project management); Language model; F1 score; Artificial intelligence; Macro; Voice activity detection; Machine learning; Sentiment analysis; Natural language processing; Speech recognition; Speech processing","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":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.001060559,0.0003562781,0.0004257763,0.0002498636,0.0002441435,0.001127894,0.005370883,0.0001543233,0.0001734165],"category_scores_gemma":[0.0005547486,0.000337981,0.000152613,0.0004490812,0.00005753564,0.001997384,0.001826394,0.0003575405,0.000313584],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008965786,"about_ca_system_score_gemma":0.000505254,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001316985,"about_ca_topic_score_gemma":0.001538926,"domain_scores_codex":[0.9960896,0.0001441473,0.0005307304,0.001838618,0.0009547873,0.0004421219],"domain_scores_gemma":[0.9944152,0.0001573052,0.0002683927,0.004194847,0.000786818,0.0001774826],"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":[0.0000904148,0.0002264308,0.000003005316,0.00002539107,0.001247007,0.0001057911,0.00005696703,0.0002541385,0.03219958,0.01231429,0.01056951,0.9429075],"study_design_scores_gemma":[0.0006123813,0.00007173966,0.00001352595,0.00003792109,0.0002583088,0.0000966681,0.00002710992,0.8471058,0.07703122,0.005822099,0.068514,0.0004092344],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0007809468,0.00009950163,0.9809421,0.003096815,0.007567625,0.0003051994,0.005000241,0.0001392384,0.002068286],"genre_scores_gemma":[0.8213546,0.001146945,0.141132,0.002067026,0.003355959,0.0002604085,0.01915976,0.00008467065,0.01143865],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9424983,"threshold_uncertainty_score":0.999909,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4481507708274435,"score_gpt":0.3651251625251364,"score_spread":0.0830256083023071,"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."}}