{"id":"W2520723151","doi":"10.7287/peerj.preprints.2373v1","title":"Stopping duplicate bug reports before they start with Continuous Querying for bug reports","year":2016,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Data deduplication; Search engine indexing; Software bug; BitTorrent tracker; Security bug; Information retrieval; Software; Process (computing); Database; Programming language; Artificial intelligence; Cloud computing","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":[0.02123406,0.001658433,0.002988259,0.005908777,0.001234236,0.004799328,0.003799966,0.002243667,0.002588733],"category_scores_gemma":[0.1285336,0.0009961155,0.001102595,0.003675981,0.001609721,0.007644271,0.003661198,0.002557806,0.001987164],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009393109,"about_ca_system_score_gemma":0.002623457,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001835295,"about_ca_topic_score_gemma":0.001824615,"domain_scores_codex":[0.9784473,0.005896398,0.003069079,0.003349463,0.008471,0.0007668451],"domain_scores_gemma":[0.8067645,0.1006986,0.02602271,0.04243496,0.0215485,0.002530657],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001454428,0.0009145762,0.05702697,0.001906943,0.0003240544,0.0006470889,0.004318655,0.005466007,0.05135108,0.00611398,0.01386213,0.8566141],"study_design_scores_gemma":[0.001152233,0.008047968,0.1397797,0.001446069,0.001674316,0.009898221,0.007600933,0.3201671,0.3509107,0.03978027,0.1182163,0.001326167],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3167354,0.005159637,0.6229573,0.002951275,0.0006053594,0.001746368,0.001555999,0.04088952,0.007399052],"genre_scores_gemma":[0.5562947,0.0008392986,0.4331017,0.0007578951,0.0003040048,0.0003139701,0.002144308,0.001998539,0.004245578],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02123406,"threshold_uncertainty_score":0.1122978,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01488962674465569,"score_gpt":0.2471598646030495,"score_spread":0.2322702378583938,"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."}}