{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0004320641,0.0003148866,0.0003323541,0.0002534663,0.0001346994,0.0001231556,0.001185646,0.00060168,0.000009978671],"category_scores_gemma":[0.000183656,0.0003783728,0.0001565141,0.0008174135,0.00006864382,0.0003004616,0.0009188229,0.001934182,0.00006457105],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004300425,"about_ca_system_score_gemma":0.000357051,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001178636,"about_ca_topic_score_gemma":0.0003592144,"domain_scores_codex":[0.9976724,0.0004436114,0.0002446891,0.001136738,0.0001575507,0.0003449451],"domain_scores_gemma":[0.9985864,0.0001335226,0.0002289897,0.0006480828,0.000209139,0.0001938991],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001218706,0.001728993,0.001100483,0.0008318696,0.0007596478,0.005716221,0.0102123,0.7694561,0.04793492,0.05881544,0.002214915,0.1000104],"study_design_scores_gemma":[0.001190439,0.0003760103,0.001043389,0.0001943126,0.0000509989,0.00001416862,0.0003778742,0.9400585,0.02381949,0.03053433,0.001418456,0.0009219869],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1005194,0.00006390183,0.8971207,0.0006702681,0.0007072338,0.0003926615,0.00001853765,0.0003005057,0.0002067355],"genre_scores_gemma":[0.9943538,0.0004144035,0.004397102,0.0003921483,0.00006992344,0.000001950806,0.00002133783,0.00001903838,0.0003303303],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8938343,"threshold_uncertainty_score":0.9998668,"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."}}