{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008953735,0.0001640138,0.0003056857,0.0002377757,0.0001873064,0.0006434995,0.0003349954,0.0001020086,8.489351e-7],"category_scores_gemma":[0.0005247981,0.0001324895,0.0001917119,0.0003084129,0.00003329122,0.001582525,0.0001022179,0.0003363427,0.000001173872],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002898531,"about_ca_system_score_gemma":0.0001223226,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001728548,"about_ca_topic_score_gemma":0.000001812092,"domain_scores_codex":[0.9983874,0.00005732349,0.000461367,0.0002908891,0.0004915133,0.0003115278],"domain_scores_gemma":[0.997772,0.001533807,0.0001382885,0.0002016045,0.0002346017,0.000119669],"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":[0.0003682704,0.0002332531,0.08067465,0.01395216,0.003400448,0.001970445,0.03596694,0.5895491,0.0007096601,0.1483834,0.04225665,0.08253507],"study_design_scores_gemma":[0.001733887,0.001555063,0.002670377,0.005532665,0.0001653734,0.003268831,0.001324105,0.9572342,0.00005410563,0.01902607,0.006751321,0.00068402],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0142804,0.009255013,0.9735868,0.0001349859,0.002253753,0.0001590134,0.00003821109,0.0002835575,0.000008274217],"genre_scores_gemma":[0.9557837,0.0000922294,0.0435165,0.00001938026,0.0003963591,0.00001817069,0.00000250071,0.00003482245,0.0001363226],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9415033,"threshold_uncertainty_score":0.6205281,"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."}}