{"id":"W4376606580","doi":"10.1109/saner56733.2023.00013","title":"Towards Understanding the Impacts of Textual Dissimilarity on Duplicate Bug Report Detection","year":2023,"lang":"en","type":"article","venue":"","topic":"Software Engineering Research","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Software bug; Visualization; Construct (python library); Software regression; Security bug; Overhead (engineering); Domain (mathematical analysis); Software; Information retrieval; Data mining; Software development; Software quality","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":[0.01123504,0.00111483,0.001180737,0.007443851,0.0009038153,0.00296187,0.001387414,0.001488465,0.0006614455],"category_scores_gemma":[0.144795,0.0006613724,0.0009003476,0.003907636,0.001347572,0.007443038,0.002363088,0.001797388,0.0003845372],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007931126,"about_ca_system_score_gemma":0.0009742513,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00304249,"about_ca_topic_score_gemma":0.004013136,"domain_scores_codex":[0.9867857,0.005459252,0.001487931,0.002425548,0.003435557,0.0004060404],"domain_scores_gemma":[0.8037245,0.1379657,0.02757635,0.0131561,0.01580626,0.001771022],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001520727,0.001053234,0.5254158,0.003005914,0.0007061848,0.001058637,0.007360207,0.01838531,0.04732759,0.002438475,0.004923502,0.3868044],"study_design_scores_gemma":[0.0002482235,0.002462843,0.496647,0.0004963152,0.00101854,0.00315075,0.006286767,0.4129354,0.05348958,0.01150281,0.01141514,0.0003467619],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9157696,0.002611936,0.07681176,0.0006637998,0.00007726952,0.0002266348,0.0009124015,0.001509429,0.00141715],"genre_scores_gemma":[0.9338094,0.0004627687,0.06318438,0.0001466152,0.00008615643,0.0001001195,0.001717967,0.0001739794,0.0003185065],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01123504,"threshold_uncertainty_score":0.05941731,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07026126813592734,"score_gpt":0.3222827029850321,"score_spread":0.2520214348491048,"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."}}