{"id":"W4389685461","doi":"10.1007/s10664-023-10409-5","title":"Detection and evaluation of bias-inducing features in machine learning","year":2023,"lang":"en","type":"article","venue":"Empirical Software Engineering","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Consortium de Recherche et d’innovation en Aérospatiale au Québec","keywords":"Computer science; Machine learning; Identification (biology); Context (archaeology); Artificial intelligence; Harm; Feature (linguistics); Outcome (game theory); Data mining; Psychology","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001276202,0.00009475324,0.0001262648,0.0003132207,0.00005351029,0.00004360809,0.000153604,0.00006322133,0.000003972247],"category_scores_gemma":[0.00265992,0.00009644419,0.00002698806,0.001178682,0.00001016876,0.0002744215,0.0001277381,0.0002284976,0.00001010746],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008251705,"about_ca_system_score_gemma":0.00002340974,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001364041,"about_ca_topic_score_gemma":0.00006239272,"domain_scores_codex":[0.998922,0.00008067436,0.0001990834,0.0002408342,0.0003398687,0.0002175757],"domain_scores_gemma":[0.9992569,0.0004322999,0.00004235404,0.0001462985,0.00007379539,0.00004835294],"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":[0.000003612331,0.00001642309,0.06790005,0.00005118032,0.000008666393,0.00001118644,0.002319871,0.7829444,0.007442107,0.0001233763,0.00001871874,0.1391604],"study_design_scores_gemma":[0.00007785369,0.00003555381,0.1341486,0.0000406902,0.000003474758,0.000004957571,0.00004046899,0.8437666,0.02133206,0.0002966005,0.0001513462,0.000101791],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7842833,0.0002115716,0.2148396,0.0001034893,0.0001405701,0.00009767535,2.657974e-7,0.0003107372,0.00001278811],"genre_scores_gemma":[0.9958068,0.00001339494,0.004091706,0.00001451201,0.00002826056,0.00001665669,0.000001438614,0.00001039787,0.00001680404],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2115235,"threshold_uncertainty_score":0.393288,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06541519331576667,"score_gpt":0.314828704639924,"score_spread":0.2494135113241573,"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."}}