{"id":"W2086095861","doi":"10.1109/tr.2012.2183912","title":"Evaluating Stratification Alternatives to Improve Software Defect Prediction","year":2012,"lang":"en","type":"article","venue":"IEEE Transactions on Reliability","topic":"Software Engineering Research","field":"Computer Science","cited_by":59,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Machine learning; Software bug; Skewness; Software; Artificial intelligence; Software quality; Data mining; Sampling (signal processing); Stratification (seeds); Reliability engineering; Statistics; Software development; Engineering; Mathematics","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":[],"consensus_categories":[],"category_scores_codex":[0.001504367,0.0001660267,0.0001334606,0.0001760745,0.0002099058,0.00009343628,0.0004632093,0.00008172872,0.00002382526],"category_scores_gemma":[0.0005723611,0.0001630266,0.0001190148,0.0005614376,0.00003548551,0.000784971,0.000004326296,0.0003429035,0.0001292601],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003263085,"about_ca_system_score_gemma":0.00007762797,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004521589,"about_ca_topic_score_gemma":0.000003155136,"domain_scores_codex":[0.9979773,0.0001824767,0.0002852048,0.0005254359,0.0006096846,0.0004198568],"domain_scores_gemma":[0.9974415,0.001056748,0.00004603973,0.0009988637,0.0002064704,0.0002503636],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00006997232,0.001061097,0.008246226,0.0001308705,0.00007858813,8.643339e-7,0.002951031,0.2681512,0.02834665,0.0003220376,0.0001976315,0.6904438],"study_design_scores_gemma":[0.001313528,0.002246769,0.3078886,0.0001371756,0.00007664832,0.00001975316,0.00009887479,0.2040105,0.4804985,0.002151433,0.0004447349,0.001113448],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2415911,0.00001430463,0.7558721,0.0001463702,0.001278189,0.0004861313,0.00002725562,0.000548736,0.00003573275],"genre_scores_gemma":[0.9073349,0.000002668651,0.0921675,0.000036092,0.00008179295,0.0002826996,0.000001876336,0.00001242358,0.00008000463],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.6893303,"threshold_uncertainty_score":0.6648033,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04061733545592563,"score_gpt":0.3384401637331717,"score_spread":0.2978228282772461,"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."}}