{"id":"W4403600269","doi":"10.1016/j.jss.2024.112266","title":"Refining software defect prediction through attentive neural models for code understanding","year":2024,"lang":"en","type":"article","venue":"Journal of Systems and Software","topic":"Software Engineering Research","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Alberta","funders":"","keywords":"Refining (metallurgy); Computer science; Code (set theory); Software; Artificial intelligence; Artificial neural network; Software bug; Machine learning; Software engineering; Programming language; Chemistry","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.000696017,0.0008803378,0.0005719523,0.0008677548,0.0002580126,0.001039823,0.001607531,0.001208687,0.00222],"category_scores_gemma":[0.005628517,0.0005170499,0.0006934223,0.0004160128,0.0004154306,0.002472209,0.0008188678,0.002222002,0.0004781432],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000937696,"about_ca_system_score_gemma":0.0008586666,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01062794,"about_ca_topic_score_gemma":0.01758582,"domain_scores_codex":[0.9997123,0.00005560597,0.00001610393,0.0001217121,0.00005048786,0.00004372152],"domain_scores_gemma":[0.9971231,0.001901068,0.0002333335,0.000239633,0.0004284073,0.00007451515],"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":[0.0004981186,0.0008006462,0.01233382,0.0001587984,0.0001686795,0.0002379183,0.0002189747,0.6233373,0.01606262,0.004569876,0.003818121,0.3377951],"study_design_scores_gemma":[0.000003176789,0.00002005477,0.0003828413,0.000004409647,0.000009358596,0.000007088629,0.000005968774,0.9967018,0.0008876111,0.001917735,0.00005691336,0.000003033002],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4362002,0.0008139951,0.5530028,0.0008857347,0.0001430987,0.0001019769,0.0004392447,0.00389917,0.004513841],"genre_scores_gemma":[0.9582016,0.0001303425,0.03990252,0.0001277021,0.00002914395,0.00003027972,0.0002845251,0.00007190498,0.001221867],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01062794,"threshold_uncertainty_score":0.02113211,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08898464247298289,"score_gpt":0.2992358029667185,"score_spread":0.2102511604937356,"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."}}