{"id":"W4383913721","doi":"10.1111/coin.12593","title":"Online short text clustering using infinite extensions of discrete mixture models","year":2023,"lang":"en","type":"article","venue":"Computational Intelligence","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Cluster analysis; Dirichlet distribution; Computer science; Multinomial distribution; Gibbs sampling; Prior probability; Artificial intelligence; Latent Dirichlet allocation; Benchmark (surveying); Outlier; Topic model; Pattern recognition (psychology); Machine learning; Mathematics; Statistics; Bayesian probability","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":[0.003012436,0.001254782,0.001932348,0.003461578,0.0009897683,0.002026238,0.003385151,0.001830376,0.002699554],"category_scores_gemma":[0.01164344,0.0008878849,0.001715486,0.002877185,0.001146897,0.003883722,0.001780895,0.002374294,0.001762751],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001684088,"about_ca_system_score_gemma":0.001119508,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009195105,"about_ca_topic_score_gemma":0.01062388,"domain_scores_codex":[0.9974627,0.001030007,0.0001392449,0.00065096,0.0005614468,0.0001555026],"domain_scores_gemma":[0.9931067,0.004540367,0.0005046023,0.000759361,0.0008282469,0.0002607616],"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":[0.0004830103,0.0002205771,0.00301634,0.0001867816,0.0002123277,0.0001885671,0.0002996712,0.7318591,0.004545347,0.0419714,0.004892451,0.2121244],"study_design_scores_gemma":[0.000004984995,0.000005898402,0.00009307633,0.000003537268,0.000003735629,0.00001120526,0.000006564298,0.9922517,0.0003234444,0.00698302,0.0003048278,0.000007978598],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01104044,0.0003112936,0.9869735,0.0001722302,0.00004078738,0.00004447068,0.0001200611,0.0008482761,0.0004490685],"genre_scores_gemma":[0.3839901,0.0006911464,0.6054638,0.0003165627,0.0002983137,0.0003303455,0.00271257,0.0004925054,0.005704705],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009195105,"threshold_uncertainty_score":0.01828319,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1131347818749508,"score_gpt":0.3644947936831115,"score_spread":0.2513600118081607,"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."}}