{"id":"W2128799871","doi":"10.1016/j.scico.2012.08.003","title":"Studying software evolution using topic models","year":2012,"lang":"en","type":"article","venue":"Science of Computer Programming","topic":"Software Engineering Research","field":"Computer Science","cited_by":94,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal; Queen's University","funders":"","keywords":"Computer science; Structuring; Topic model; Software evolution; Source code; Software; Software development; Data science; Software maintenance; Task (project management); Software system; Code (set theory); Generative grammar; Software engineering; Information retrieval; Software construction; Artificial intelligence; Programming language","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":[0.001673375,0.0001432061,0.00017617,0.0004425485,0.0002822871,0.0002457023,0.001615372,0.00004197039,0.000001274906],"category_scores_gemma":[0.0001749741,0.0001390728,0.00006089491,0.001934137,0.0002512696,0.003214737,0.00100372,0.0001561351,0.000005197834],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002333457,"about_ca_system_score_gemma":0.0002330981,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003846066,"about_ca_topic_score_gemma":2.972256e-7,"domain_scores_codex":[0.9974102,0.00003868389,0.0002631552,0.000393723,0.001000424,0.0008937712],"domain_scores_gemma":[0.9985911,0.000178566,0.00009417332,0.0006277136,0.0002795262,0.0002289813],"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":[0.000002982695,0.0003223284,0.105017,0.0001460049,0.00002552077,0.000005500071,0.005050587,0.07523499,0.004489928,0.052301,0.00001985403,0.7573843],"study_design_scores_gemma":[0.0002088742,0.0001069702,0.02251228,0.0001052799,0.000005436609,0.00004008131,0.00004756897,0.9721562,0.002609152,0.001784589,0.0001116113,0.0003119327],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2825828,0.0002123253,0.7160724,0.00001497137,0.0006991751,0.0001706267,1.386113e-7,0.0002370944,0.00001046275],"genre_scores_gemma":[0.5347418,4.296664e-7,0.4651445,0.000005138293,0.00009451321,0.000005081673,1.072582e-7,0.000005284755,0.000003134388],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8969212,"threshold_uncertainty_score":0.5671225,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05347133154518463,"score_gpt":0.3013226588468253,"score_spread":0.2478513273016407,"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."}}