{"id":"W3158220826","doi":"10.1177/20597991211010416","title":"Big data and democratic speech: Predicting deliberative quality using machine learning techniques","year":2021,"lang":"en","type":"article","venue":"Methodological Innovations","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Newspaper; Quality (philosophy); Deliberative democracy; Computer science; Legislation; Democracy; Data science; Artificial intelligence; Machine learning; Political science; Law; Politics; Epistemology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.01229721,0.0009069869,0.0008099438,0.008383473,0.001359372,0.003683897,0.001345718,0.001615906,0.001421078],"category_scores_gemma":[0.06362283,0.0004562941,0.001025187,0.006141822,0.001755725,0.003521939,0.002320994,0.002882523,0.0006288385],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001698334,"about_ca_system_score_gemma":0.001267528,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008417306,"about_ca_topic_score_gemma":0.01195991,"domain_scores_codex":[0.9910179,0.005985272,0.0007121928,0.0008462187,0.001069164,0.000369343],"domain_scores_gemma":[0.8906537,0.08719129,0.009799161,0.006756706,0.004250663,0.001348559],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003680283,0.0008191561,0.8084003,0.0003027122,0.0006757857,0.0002490249,0.002905712,0.05967845,0.001006935,0.007687259,0.00499657,0.1129101],"study_design_scores_gemma":[0.00004381356,0.0001391403,0.2358526,0.0001920035,0.00009864419,0.0001531922,0.003926981,0.7087196,0.001924582,0.0431022,0.005729296,0.0001179249],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8144792,0.001311204,0.1659517,0.005307934,0.0001757613,0.0004371564,0.00539612,0.0004982052,0.006442793],"genre_scores_gemma":[0.9631516,0.0001565303,0.0332098,0.0001181057,0.0001132484,0.0002189276,0.002599113,0.0000312096,0.0004014645],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01229721,"threshold_uncertainty_score":0.06503463,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.5102523518130553,"score_gpt":0.4385821259810788,"score_spread":0.07167022583197652,"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."}}