{"id":"W2612562696","doi":"10.1109/icit.2017.7915513","title":"Unsupervised learning of finite mixtures using scaled dirichlet distribution and its application to software modules categorization","year":2017,"lang":"en","type":"article","venue":"","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":46,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Dirichlet distribution; Mixture model; Computer science; Expectation–maximization algorithm; Latent Dirichlet allocation; Artificial intelligence; Flexibility (engineering); Machine learning; Dirichlet process; Data mining; Pattern recognition (psychology); Algorithm; Topic model; Mathematics; Maximum likelihood; Statistics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006290048,0.0009863874,0.002103058,0.003292636,0.001435909,0.002104049,0.003207699,0.002257116,0.001777246],"category_scores_gemma":[0.0186811,0.001001807,0.002733916,0.00213772,0.001950538,0.003387434,0.002501938,0.00331119,0.0007658121],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002008907,"about_ca_system_score_gemma":0.001594456,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006283322,"about_ca_topic_score_gemma":0.006517313,"domain_scores_codex":[0.9960509,0.001994048,0.0002275708,0.0008903894,0.0006121699,0.0002248393],"domain_scores_gemma":[0.9903638,0.007435112,0.0004116737,0.0007044426,0.0008539614,0.0002310425],"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":[0.0003066076,0.000290911,0.004375995,0.0002026987,0.0003032788,0.0001715349,0.0008074279,0.6207023,0.00462195,0.06065895,0.003558979,0.3039993],"study_design_scores_gemma":[0.000007033482,0.00001010569,0.0001683857,0.000007927022,0.0000062596,0.00002564841,0.00001830956,0.9777722,0.0006967725,0.02092504,0.0003479262,0.0000143394],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006350807,0.0001061931,0.9927757,0.0001393819,0.00001910148,0.00004229137,0.00003488388,0.0003179937,0.000213727],"genre_scores_gemma":[0.2621771,0.0002609738,0.7343051,0.0002668573,0.0001230026,0.0003660995,0.0006186526,0.000248574,0.001633715],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006290048,"threshold_uncertainty_score":0.03326535,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02167367406103386,"score_gpt":0.2850443828201396,"score_spread":0.2633707087591057,"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."}}