{"id":"W1999910737","doi":"10.1109/msr.2013.6624026","title":"A contextual approach towards more accurate duplicate bug report detection","year":2013,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":129,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Data deduplication; Software bug; BitTorrent tracker; Security bug; Software regression; Vocabulary; Software; Context (archaeology); Android (operating system); Debugging; Data science; World Wide Web; Software engineering; Software quality; Software development; Artificial intelligence; Database; Computer security; Software security assurance; Information security; Eye tracking","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":[0.005642478,0.001435877,0.002447438,0.01046106,0.001607362,0.002794836,0.001946378,0.001475761,0.001987924],"category_scores_gemma":[0.03350873,0.0008369978,0.001163809,0.006908378,0.001011206,0.004384126,0.00390643,0.001661125,0.001334476],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008452418,"about_ca_system_score_gemma":0.002340989,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004575371,"about_ca_topic_score_gemma":0.01010195,"domain_scores_codex":[0.993145,0.002017511,0.0008258612,0.002085721,0.001587348,0.0003384322],"domain_scores_gemma":[0.9709781,0.009129726,0.003530783,0.009070032,0.006727538,0.000563856],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006559823,0.0005484193,0.02953391,0.001497032,0.0002732697,0.0007548867,0.002765972,0.01330263,0.0488185,0.009191772,0.01667841,0.8759793],"study_design_scores_gemma":[0.0003271593,0.001324224,0.05476442,0.000766136,0.001383554,0.004553251,0.003849331,0.6219202,0.1248781,0.04743015,0.138126,0.000677459],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07060857,0.005923462,0.9025314,0.001225271,0.0003836436,0.0004653271,0.001632264,0.01302176,0.004208346],"genre_scores_gemma":[0.2942924,0.001252369,0.6984233,0.0004081794,0.0005403596,0.0003175983,0.002355132,0.0006356477,0.001775021],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01046106,"threshold_uncertainty_score":0.02984059,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02898033520437638,"score_gpt":0.277534679319666,"score_spread":0.2485543441152896,"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."}}