{"id":"W2142978025","doi":"10.1145/1858996.1859015","title":"Deviance from perfection is a better criterion than closeness to evil when identifying risky code","year":2010,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":63,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Perfection; Code (set theory); Computer science; Closeness; Detector; Normality; Deviance (statistics); Measure (data warehouse); Set (abstract data type); Robustness (evolution); Algorithm; Artificial intelligence; Theoretical computer science; Computer security; Machine learning; Programming language; Data mining; Mathematics; Statistics; Epistemology","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":[0.0003434278,0.0001446329,0.0001359031,0.0001582028,0.000122828,0.0005094329,0.0009019607,0.00008957197,0.0002242067],"category_scores_gemma":[0.0002494583,0.00013837,0.00005579854,0.0003290938,0.00001881448,0.0006797827,0.0004087388,0.0003402698,0.0006338586],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004902988,"about_ca_system_score_gemma":0.0000321641,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008760996,"about_ca_topic_score_gemma":0.0004525922,"domain_scores_codex":[0.998449,0.00004098006,0.0001676337,0.0005525383,0.0004374636,0.000352396],"domain_scores_gemma":[0.9985718,0.0002950859,0.00002649577,0.0008234575,0.0001209446,0.0001622733],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.00001925548,0.0001035424,0.3917333,0.00009746096,0.00006178132,0.00005291667,0.007339114,0.00008530677,0.3846795,0.001023005,0.01461874,0.200186],"study_design_scores_gemma":[0.0004508875,0.00008838972,0.7220966,0.0001419575,0.000009145165,0.00002584315,0.00002088398,0.03660485,0.2017739,0.002868572,0.03522446,0.0006945301],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5257471,0.0000217441,0.471858,0.001148063,0.0008120392,0.00009711851,0.000003126622,0.0002723334,0.00004056283],"genre_scores_gemma":[0.869247,0.000004498565,0.129072,0.0007264986,0.0002896785,0.0000375542,0.000002033904,0.00001913142,0.0006015536],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3435,"threshold_uncertainty_score":0.8147182,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02695663459996754,"score_gpt":0.2989857409719689,"score_spread":0.2720291063720013,"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."}}