{"id":"W2967758098","doi":"10.1145/3338906.3341182","title":"CloneCognition: machine learning based code clone validation tool","year":2019,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Saskatchewan","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund","keywords":"clone (Java method); Computer science; Source code; Code (set theory); Generalization; Software maintenance; Programming language; Program comprehension; Software; Codebase; Software system; Process (computing); Artificial intelligence; Software engineering; Machine learning; Set (abstract data type)","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0004137335,0.00008887182,0.00009256233,0.0001248285,0.00005471893,0.00014564,0.0003862832,0.00004342549,0.0006377042],"category_scores_gemma":[0.0003415106,0.00008484221,0.00003693211,0.0003690964,0.000008592183,0.0003563848,0.000120107,0.0002067949,0.001595691],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004210571,"about_ca_system_score_gemma":0.00004618481,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002130437,"about_ca_topic_score_gemma":0.000001132652,"domain_scores_codex":[0.9989458,0.00005324161,0.0001279695,0.0002830416,0.0003647723,0.0002251998],"domain_scores_gemma":[0.9989024,0.0005568069,0.00002583898,0.0003662174,0.00009395251,0.00005479001],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006643502,0.0004258532,0.5911626,0.0003789294,0.0001006172,0.00009468178,0.0004847504,0.209381,0.04271672,0.02626957,0.00408156,0.1248373],"study_design_scores_gemma":[0.0006706176,0.0001314178,0.02399646,0.00002088264,0.000002029897,0.000008553165,0.000002899301,0.9431136,0.02582568,0.0001814969,0.005818745,0.0002276219],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2120486,0.00002128383,0.7854965,0.0004516313,0.0002065334,0.0001512249,0.000001351852,0.000675494,0.0009472998],"genre_scores_gemma":[0.9486777,0.000002669728,0.04888112,0.000140296,0.00003566264,0.00001352601,0.00002468427,0.0000119033,0.002212424],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7366291,"threshold_uncertainty_score":0.9991817,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01768197075489549,"score_gpt":0.2522891837512765,"score_spread":0.234607212996381,"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."}}