{"id":"W4389980560","doi":"10.1016/j.jss.2023.111934","title":"A survey on machine learning techniques applied to source code","year":2023,"lang":"en","type":"article","venue":"Journal of Systems and Software","topic":"Software Engineering Research","field":"Computer Science","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University; Dalhousie University","funders":"H2020 European Research Council; European Research Council","keywords":"Computer science; Workflow; Source code; Machine learning; Context (archaeology); Software; Software engineering; Artificial intelligence; Task (project management); Code review; Data science; Static program analysis; Software development; Systems engineering; Programming language; Engineering; Database","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":[0.002142751,0.0001113213,0.0002645488,0.0004285166,0.0001012072,0.0002083397,0.00045528,0.00005897749,0.000001098953],"category_scores_gemma":[0.001371163,0.00008861208,0.00003867232,0.0006335604,0.00001162313,0.0000979564,0.0001783308,0.000355387,0.00002608332],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004950494,"about_ca_system_score_gemma":0.00004290068,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009212895,"about_ca_topic_score_gemma":0.000006009231,"domain_scores_codex":[0.9986948,0.0000960897,0.0002970185,0.0001717332,0.0005018225,0.0002385836],"domain_scores_gemma":[0.9981806,0.001131185,0.0001189711,0.0002092374,0.0001719634,0.0001880769],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0001919192,0.0001369833,0.6113157,0.0006737718,0.0002555404,0.0006061586,0.005819876,0.07850816,0.002891057,0.001091472,0.03507026,0.2634391],"study_design_scores_gemma":[0.001362554,0.002642252,0.8299409,0.001727834,0.00001504753,0.0005382153,0.000263106,0.02642163,0.001963906,0.0001466269,0.1339205,0.001057464],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2025485,0.0004152389,0.7956157,0.0001831627,0.0003718509,0.0002212149,0.00000873165,0.0006158195,0.00001979665],"genre_scores_gemma":[0.9921591,0.00004643124,0.006878717,0.00004500593,0.0001454968,0.00001001214,0.000001897659,0.00002492705,0.0006884705],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7896106,"threshold_uncertainty_score":0.3613496,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02886875562047403,"score_gpt":0.2772920157214117,"score_spread":0.2484232601009377,"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."}}