{"id":"W4389042830","doi":"10.1109/isncc58260.2023.10323873","title":"Generalized Probabilistic Clustering Projection Models for Discrete Data","year":2023,"lang":"en","type":"article","venue":"","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Latent Dirichlet allocation; Cluster analysis; Perplexity; Prior probability; Dirichlet distribution; Projection (relational algebra); Mathematics; Pattern recognition (psychology); Generalized Dirichlet distribution; Computer science; Artificial intelligence; Probabilistic logic; Topic model; Algorithm; Bayesian probability; Dirichlet series; Language model","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":[0.007822529,0.001755358,0.002764734,0.00263884,0.001226602,0.003375276,0.004134648,0.002961472,0.005239435],"category_scores_gemma":[0.0261884,0.001477548,0.002705931,0.004122271,0.002945383,0.005298385,0.003492621,0.004933945,0.001858369],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002881906,"about_ca_system_score_gemma":0.002205013,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01125683,"about_ca_topic_score_gemma":0.01302815,"domain_scores_codex":[0.9934673,0.003681217,0.0002766456,0.001384843,0.0008954751,0.000294539],"domain_scores_gemma":[0.9844525,0.0117516,0.0007839286,0.00147859,0.001273514,0.0002598479],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001422794,0.00008037704,0.00114313,0.0002420708,0.0001814708,0.0001192676,0.0005197313,0.691204,0.0008082514,0.2443894,0.002720139,0.05844989],"study_design_scores_gemma":[0.00001162083,0.00001008686,0.0001437359,0.0000166285,0.000009476918,0.00002665854,0.00002497824,0.888003,0.0001479483,0.1105946,0.0009929168,0.00001842279],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00435438,0.0002873496,0.9938251,0.0003026606,0.00002800147,0.00006405922,0.0001746311,0.0002771548,0.0006867787],"genre_scores_gemma":[0.2685985,0.001467952,0.7151873,0.0004571249,0.0002235276,0.001265923,0.002047621,0.0003815663,0.01037053],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01125683,"threshold_uncertainty_score":0.04136997,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1401373378521512,"score_gpt":0.3548434531343582,"score_spread":0.2147061152822071,"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."}}