{"id":"W2054765427","doi":"10.1007/s10115-011-0467-4","title":"A countably infinite mixture model for clustering and feature selection","year":2011,"lang":"en","type":"article","venue":"Knowledge and Information Systems","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":52,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université de Sherbrooke; Concordia University","funders":"","keywords":"Mixture model; Cluster analysis; Model selection; Computer science; Artificial intelligence; Feature selection; Dirichlet process; Dirichlet distribution; Machine learning; Bayesian inference; Inference; Pattern recognition (psychology); Data mining; Mathematics; Bayesian probability","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":[0.008247054,0.001524198,0.003856428,0.003741367,0.002002191,0.004424692,0.008513694,0.004212843,0.005240204],"category_scores_gemma":[0.02975276,0.002046646,0.003454539,0.006218964,0.003559905,0.007841675,0.003938571,0.005627663,0.002222069],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00303307,"about_ca_system_score_gemma":0.002081005,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009596976,"about_ca_topic_score_gemma":0.009844458,"domain_scores_codex":[0.9933289,0.003547089,0.0003676968,0.001171602,0.001279784,0.0003049647],"domain_scores_gemma":[0.9853056,0.01111777,0.0006226068,0.001342307,0.001295963,0.0003157322],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000151014,0.00009760749,0.0006613346,0.0002714892,0.0002430649,0.0001179351,0.0002921236,0.274262,0.0008596445,0.6372823,0.004785927,0.0809755],"study_design_scores_gemma":[0.00001483197,0.00001197374,0.0001203612,0.00002649153,0.00003148201,0.00004537527,0.00001375532,0.7168162,0.0001630967,0.2811109,0.001612263,0.00003313991],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0008769478,0.0003146155,0.9981573,0.0001685873,0.00001681386,0.00001489428,0.00005999161,0.00006813517,0.0003227085],"genre_scores_gemma":[0.120834,0.001724557,0.8660718,0.000483067,0.0003678806,0.0007573895,0.001223373,0.0002855387,0.008252461],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009596976,"threshold_uncertainty_score":0.0436151,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02864478443536663,"score_gpt":0.2546671348517391,"score_spread":0.2260223504163725,"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."}}