{"id":"W4250023757","doi":"10.1109/icse.2015.91","title":"Revisiting the Impact of Classification Techniques on the Performance of Defect Prediction Models","year":2015,"lang":"en","type":"article","venue":"2015 IEEE/ACM 37th IEEE International Conference on Software Engineering","topic":"Software Engineering Research","field":"Computer Science","cited_by":319,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Replicate; Machine learning; Predictive modelling; Artificial intelligence; Software bug; Software; Data mining; Multivariate adaptive regression splines; Multivariate statistics; Set (abstract data type); Variety (cybernetics); Support vector machine; Regression; Logistic regression; Regression analysis; Statistics; Mathematics","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":[0.001312997,0.0002407809,0.0002240296,0.0002966157,0.00006034155,0.0001018818,0.00208708,0.00009692758,0.000009683381],"category_scores_gemma":[0.002173821,0.0001573184,0.0001577583,0.0004169593,0.00006297287,0.0004979611,0.0001425533,0.0004621503,0.0000132088],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003170451,"about_ca_system_score_gemma":0.0002246785,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004507759,"about_ca_topic_score_gemma":2.735293e-7,"domain_scores_codex":[0.9978719,0.00007270344,0.000442996,0.0003349767,0.001001404,0.0002760631],"domain_scores_gemma":[0.9966825,0.001099281,0.0002314303,0.001016321,0.0008731777,0.00009725388],"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":[0.0001151903,0.0001117084,0.0275577,0.0001135927,0.0003121638,0.000004324661,0.001212682,0.9074306,0.01174616,0.03395075,0.002268325,0.01517677],"study_design_scores_gemma":[0.0001719534,0.0003256787,0.01578472,0.0004657571,0.00000679243,0.00001206082,0.00002291024,0.9639622,0.01866093,0.0003638548,0.00004924191,0.0001739028],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.574836,0.00003720929,0.4231385,0.0003461652,0.0006645048,0.0003286738,0.0000315723,0.0003663841,0.0002509897],"genre_scores_gemma":[0.9914337,0.00004864248,0.008095773,0.00001509582,0.0002476297,0.00008565762,0.000007163043,0.00002589894,0.00004037111],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4165978,"threshold_uncertainty_score":0.641526,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1198000391827528,"score_gpt":0.3409680431586617,"score_spread":0.2211680039759089,"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."}}