{"id":"W3094997061","doi":"10.4224/40002027","title":"Towards Ethics by Design in Online Abusive Content Detection","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"","keywords":"Inclusion (mineral); Task (project management); Field (mathematics); Computer science; Process (computing); Identity (music); Annotation; Content analysis; Internet privacy; Data science; Computer security; Psychology; Artificial intelligence; Social psychology; Sociology; Engineering","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":[0.1310072,0.001182256,0.001445688,0.003609113,0.0050728,0.01819529,0.003622918,0.006506476,0.002511456],"category_scores_gemma":[0.2314485,0.00179966,0.001619893,0.00246509,0.03265242,0.02244341,0.01288909,0.009796843,0.001564544],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005678404,"about_ca_system_score_gemma":0.008739971,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001422435,"about_ca_topic_score_gemma":0.001469047,"domain_scores_codex":[0.8392784,0.1329236,0.00438552,0.00871079,0.01294522,0.001756455],"domain_scores_gemma":[0.7131413,0.1905,0.01869833,0.05326546,0.02030151,0.004093417],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001246063,0.0002263268,0.01391122,0.0003937521,0.00008085454,0.0002095322,0.01418571,0.007546904,0.00144963,0.8693451,0.006518271,0.08600809],"study_design_scores_gemma":[0.00002873456,0.00005849804,0.001059021,0.0002056833,0.00002577548,0.0002599854,0.001541002,0.03495257,0.001575234,0.9391075,0.02114226,0.00004370188],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01796755,0.0004429077,0.9516755,0.01973483,0.0002074877,0.0003244611,0.0001217287,0.0003339612,0.009191564],"genre_scores_gemma":[0.4154189,0.0004736851,0.5744854,0.003969992,0.0003455838,0.001391171,0.0003952199,0.0005208696,0.002999213],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.1310072,"threshold_uncertainty_score":0.6928404,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1848277258687152,"score_gpt":0.219594687501319,"score_spread":0.03476696163260379,"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."}}