{"id":"W2113772893","doi":"10.1016/j.scico.2013.03.015","title":"Using heuristics to estimate an appropriate number of latent topics in source code analysis","year":2013,"lang":"en","type":"article","venue":"Science of Computer Programming","topic":"Software Engineering Research","field":"Computer Science","cited_by":39,"is_retracted":false,"has_abstract":false,"ca_institutions":"Queen's University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Latent Dirichlet allocation; Source code; Topic model; Heuristics; Set (abstract data type); Locality; Cluster analysis; Code review; Code (set theory); KPI-driven code analysis; Software; Programming language; Theoretical computer science; Information retrieval; Static program analysis; Software development; Artificial intelligence","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001003657,0.0001362115,0.000308108,0.0006756663,0.00007145251,0.0002864291,0.001704042,0.00004022062,0.000003940841],"category_scores_gemma":[0.0001630114,0.000129945,0.00006632126,0.004758418,0.0002038267,0.0006766515,0.0007770419,0.0001300347,0.000006445217],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007878344,"about_ca_system_score_gemma":0.0001361383,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003939677,"about_ca_topic_score_gemma":0.00001307319,"domain_scores_codex":[0.997704,0.00004419931,0.0004192894,0.0005037841,0.0007760258,0.0005527551],"domain_scores_gemma":[0.9984125,0.0001022205,0.0001193302,0.0007218256,0.0004211357,0.0002229188],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00000186958,0.0001537353,0.2726586,0.00006152776,0.00002906899,0.000005310081,0.001963625,0.4365238,0.001532453,0.001427494,0.000002403166,0.2856402],"study_design_scores_gemma":[0.0000856745,0.00006503934,0.1084972,0.00003693917,0.000008524692,0.000003936011,0.000008182446,0.8895506,0.00130283,0.0002844677,0.00002358067,0.0001329635],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.483899,0.000003761399,0.5157751,0.00003393835,0.00008133837,0.000156295,2.371334e-7,0.00004722548,0.00000314642],"genre_scores_gemma":[0.5048447,1.750841e-7,0.4951167,0.000006396669,0.00001590615,0.000006065522,2.961148e-7,0.000004475394,0.000005232444],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.4530269,"threshold_uncertainty_score":0.5299001,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03221342878942683,"score_gpt":0.3421796663322614,"score_spread":0.3099662375428346,"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."}}