{"id":"W4210798472","doi":"10.1016/j.osnem.2021.100194","title":"Selecting and combining complementary feature representations and classifiers for hate speech detection","year":2022,"lang":"en","type":"article","venue":"Online Social Networks and Media","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":false,"ca_institutions":"École de Technologie Supérieure; Université du Québec à Montréal","funders":"","keywords":"Computer science; Heuristics; Sarcasm; Artificial intelligence; Feature selection; Machine learning; Feature extraction; Task (project management); Selection (genetic algorithm); Speech recognition; Natural language 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002512091,0.00180433,0.002196937,0.004052607,0.0007458005,0.001918948,0.001074688,0.001816758,0.001811414],"category_scores_gemma":[0.00516578,0.0004405521,0.001659346,0.002333213,0.0004588403,0.002071606,0.001624777,0.001367412,0.001649436],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004878076,"about_ca_system_score_gemma":0.001097985,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00339231,"about_ca_topic_score_gemma":0.004617471,"domain_scores_codex":[0.9982983,0.0003038066,0.00009407722,0.0004563473,0.000444259,0.0004031084],"domain_scores_gemma":[0.9976271,0.0009842872,0.0001299063,0.0002097476,0.0009039066,0.0001451045],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0007653005,0.001038495,0.01487279,0.0001675225,0.0003570874,0.0002232634,0.000112942,0.01045557,0.06102239,0.0009553825,0.005757495,0.9042717],"study_design_scores_gemma":[0.00008191239,0.0007598845,0.02858319,0.00007637579,0.000873688,0.0004856442,0.0004951613,0.8985497,0.06075902,0.005040651,0.004169032,0.0001257093],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2704646,0.002499271,0.7155703,0.0006948172,0.0005951515,0.0004167679,0.0009771311,0.003365078,0.005416813],"genre_scores_gemma":[0.7983771,0.0007488236,0.1938518,0.0002884863,0.0003382055,0.0002514768,0.001788757,0.0001332217,0.004222097],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004052607,"threshold_uncertainty_score":0.0132854,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0194614751732331,"score_gpt":0.2694540405599481,"score_spread":0.249992565386715,"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."}}