{"id":"W4366966833","doi":"10.1109/wi-iat55865.2022.00012","title":"Predicting Hateful Discussions on Reddit using Graph Transformer Networks and Communal Context","year":2022,"lang":"en","type":"article","venue":"","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Conversation; Computer science; Transformer; Social media; Graph; Language model; Context model; Artificial intelligence; Data science; Natural language processing; World Wide Web; Theoretical computer science; Sociology; Communication","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":[0.000306703,0.0001089799,0.0001100788,0.00009869405,0.001189807,0.0001193459,0.0003000972,0.00003262433,0.0001020292],"category_scores_gemma":[0.000006390431,0.00008592684,0.00005601056,0.0003710121,0.00003639722,0.0002429297,0.0001328606,0.0003582746,0.000002300296],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003747856,"about_ca_system_score_gemma":0.00001846858,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002223037,"about_ca_topic_score_gemma":0.00005724814,"domain_scores_codex":[0.9989663,0.0001249632,0.0001668945,0.00025555,0.000229399,0.0002568683],"domain_scores_gemma":[0.9995138,0.00006449698,0.00004547284,0.0002670222,0.00001594458,0.00009328724],"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":[0.0002170611,0.0004893724,0.009064435,0.00002625223,0.0001798221,0.00009455537,0.00594711,0.1152307,0.005869253,0.06622103,0.003388997,0.7932714],"study_design_scores_gemma":[0.0007428328,0.0005552222,0.00178965,0.00003990997,0.00001688442,0.0001883903,0.001512766,0.9858204,0.001337059,0.001466661,0.006196874,0.0003334132],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4358511,0.0001299875,0.5578706,0.00109998,0.0007769254,0.0002425681,0.00000515885,0.0002970985,0.00372666],"genre_scores_gemma":[0.9972924,0.00001358162,0.001886746,0.0004412007,0.00004283385,0.00001656296,0.000003060424,0.000008643286,0.000294966],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8705897,"threshold_uncertainty_score":0.9151157,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01596948223165954,"score_gpt":0.2265711998922252,"score_spread":0.2106017176605657,"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."}}