{"id":"W3093046707","doi":"10.1007/s10515-020-00277-4","title":"Understanding machine learning software defect predictions","year":2020,"lang":"en","type":"article","venue":"Automated Software Engineering","topic":"Software Engineering Research","field":"Computer Science","cited_by":98,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Exploit; Machine learning; Software bug; Software; Artificial intelligence; Boosting (machine learning); Source code; Set (abstract data type); Data mining; Principal (computer security); Decision tree; Predictive modelling; Programming language","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":[0.0003077537,0.0004039567,0.0003523218,0.0003425675,0.0003060109,0.0003191351,0.001058194,0.000180641,0.0000368882],"category_scores_gemma":[0.007061739,0.0004512262,0.000199179,0.001807756,0.00003033005,0.0007438899,0.000513081,0.0009288591,0.0001621844],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003805249,"about_ca_system_score_gemma":0.0001033828,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001672944,"about_ca_topic_score_gemma":0.000001054434,"domain_scores_codex":[0.9973335,0.00006239132,0.0003847802,0.0007248409,0.0006472575,0.0008472778],"domain_scores_gemma":[0.9972711,0.00154183,0.00007256218,0.0005152576,0.00008961852,0.0005096517],"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":[0.000006376208,0.00002959563,0.03943033,0.000244171,0.0001968866,0.0001874392,0.001143559,0.9544665,0.0005465361,0.001175254,0.001921398,0.0006519323],"study_design_scores_gemma":[0.0004244085,0.0001378015,0.005782936,0.00009439888,0.00001813727,0.00005818127,0.00001699647,0.9902123,0.0003683929,0.00003033988,0.002356467,0.0004996065],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006246235,0.0004687325,0.9527532,0.000292419,0.000478286,0.0002414406,0.00001260222,0.03949057,0.00001649637],"genre_scores_gemma":[0.8496999,0.00002715026,0.1497957,0.0001157582,0.000141273,0.00004781721,0.00003074511,0.0001122248,0.00002942704],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8434536,"threshold_uncertainty_score":0.9997939,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04516996975354468,"score_gpt":0.2432856168383505,"score_spread":0.1981156470848059,"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."}}