{"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":"codex-gemma-dda1882f352a","candidate_categories":["sts"],"consensus_categories":[],"category_scores_codex":[0.0002855451,0.0000928579,0.0001253262,0.00005766841,0.001384861,0.000105873,0.00007750215,0.00004754938,0.00000299939],"category_scores_gemma":[0.00002538199,0.0001018945,0.00002718399,0.0002404164,0.00004472075,0.0001372652,0.0001696978,0.0002719656,5.573893e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002764937,"about_ca_system_score_gemma":0.00001744355,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008607187,"about_ca_topic_score_gemma":0.0004432322,"domain_scores_codex":[0.9991836,0.00007199668,0.0001194936,0.0002834543,0.0001244968,0.0002169548],"domain_scores_gemma":[0.9995974,0.0001476148,0.0000799468,0.00006513522,0.00003415946,0.00007570779],"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":[0.00005484261,0.00004667219,0.00173386,0.00002043365,0.00005372671,0.000008240329,0.003473683,0.0002619679,0.0009149444,0.00133497,0.002580371,0.9895163],"study_design_scores_gemma":[0.003614753,0.0006821579,0.02841924,0.00002917863,0.0001140203,0.0002272641,0.01042167,0.9117361,0.000307596,0.005627293,0.0381713,0.0006493918],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8141102,0.0008996271,0.1766999,0.005474787,0.001727902,0.0006691669,0.00006367731,0.0002194742,0.0001352238],"genre_scores_gemma":[0.9869801,0.0001388778,0.01169834,0.0003757581,0.0005608946,0.00003972233,0.0001056646,0.00001190696,0.00008871013],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9888669,"threshold_uncertainty_score":0.9999152,"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."}}