{"id":"W2907794603","doi":"10.1007/s13042-018-0900-z","title":"Simultaneous clustering and feature selection via nonparametric Pitman–Yor process mixture models","year":2019,"lang":"en","type":"article","venue":"International Journal of Machine Learning and Cybernetics","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University","funders":"National Natural Science Foundation of China","keywords":"Dirichlet process; Computer science; Mixture model; Artificial intelligence; Machine learning; Cluster analysis; Model selection; Inference; Nonparametric statistics; Hierarchical Dirichlet process; Feature selection; Process (computing); Bayesian inference; Feature (linguistics); Generative model; Bayesian probability; Generative grammar; Latent Dirichlet allocation; Mathematics; Topic model; Econometrics","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":[0.004218762,0.001151815,0.003009233,0.001640079,0.00103369,0.001762022,0.002977703,0.002279731,0.001620119],"category_scores_gemma":[0.01244798,0.00159592,0.002756583,0.002223503,0.001276493,0.00296657,0.002793116,0.002225293,0.001041322],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006346148,"about_ca_system_score_gemma":0.001160138,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003439305,"about_ca_topic_score_gemma":0.003798818,"domain_scores_codex":[0.99762,0.0009612906,0.0001308909,0.0005250922,0.0005759683,0.0001866441],"domain_scores_gemma":[0.9952447,0.003157255,0.0003512018,0.0004960686,0.0006445511,0.0001062088],"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":[0.0007607954,0.0002037979,0.001761311,0.0003076053,0.0005246403,0.0002673157,0.0003901162,0.6209118,0.01058612,0.06145439,0.002327955,0.3005041],"study_design_scores_gemma":[0.00001275224,0.00002090967,0.0002548689,0.000004471929,0.00001820073,0.00002930547,0.000007904237,0.9911419,0.0007432823,0.007426127,0.000324743,0.00001552786],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.004856811,0.0001099714,0.9946527,0.00005312965,0.00001376189,0.00002041983,0.0000178345,0.0001125689,0.0001628172],"genre_scores_gemma":[0.3026016,0.0004064683,0.6915978,0.0001281646,0.0001378654,0.0003987268,0.0004270147,0.0002218666,0.004080506],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004218762,"threshold_uncertainty_score":0.02231121,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005642067798220262,"score_gpt":0.2623153025612442,"score_spread":0.256673234763024,"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."}}