{"id":"W4220918566","doi":"10.18280/isi.270114","title":"Detecting and Mitigating Bias in Data Using Machine Learning with Pre-Training Metrics","year":2022,"lang":"en","type":"article","venue":"Ingénierie des systèmes d information","topic":"Imbalanced Data Classification Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Metric (unit); Divergence (linguistics); Computer science; Performance metric; Sampling bias; Sample (material); Artificial intelligence; Statistics; Pattern recognition (psychology); False positive rate; Work (physics); Machine learning; Sample size determination; Mathematics; Engineering","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01029928,0.001673222,0.001764774,0.003246019,0.0008483973,0.003091077,0.001393539,0.001391248,0.000832217],"category_scores_gemma":[0.02790638,0.0004138878,0.001006717,0.002100385,0.0009106056,0.003362788,0.001915125,0.001986651,0.0008302507],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001009392,"about_ca_system_score_gemma":0.001529402,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001058471,"about_ca_topic_score_gemma":0.001310869,"domain_scores_codex":[0.992079,0.002013129,0.001054146,0.001531201,0.002889122,0.0004334795],"domain_scores_gemma":[0.9834635,0.006206004,0.002403546,0.003058152,0.00456592,0.0003029887],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005649384,0.00047856,0.04707469,0.0004587802,0.0003381339,0.0003213869,0.0005263974,0.04275174,0.04740649,0.005110627,0.003601815,0.8513665],"study_design_scores_gemma":[0.00004037115,0.0008703514,0.02522545,0.0001632055,0.0001519805,0.0009328278,0.0003933696,0.8423426,0.1076315,0.0132908,0.008847002,0.0001106596],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08967909,0.0009375361,0.9043812,0.0004614158,0.0002311747,0.0002384622,0.0001993523,0.002691508,0.001180391],"genre_scores_gemma":[0.5644052,0.0003777353,0.431957,0.0002682771,0.0001438169,0.0002926888,0.0008286338,0.000305445,0.001421256],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9897007,"threshold_uncertainty_score":0.05446845,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08069774931360683,"score_gpt":0.283910502966973,"score_spread":0.2032127536533662,"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."}}