{"id":"W4367319399","doi":"10.1016/j.eswa.2023.120262","title":"A context-enhanced Dirichlet model for online clustering in short text streams","year":2023,"lang":"en","type":"article","venue":"Expert Systems with Applications","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Cluster analysis; Latent Dirichlet allocation; Inference; Data mining; Artificial intelligence; Topic model; Exploit; Machine learning","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":[0.005896162,0.001228726,0.003548997,0.003298962,0.001732974,0.002998741,0.005051563,0.004004266,0.003735695],"category_scores_gemma":[0.02026225,0.001559277,0.002981675,0.004727568,0.00146525,0.0051793,0.003397407,0.004373997,0.002797948],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001954485,"about_ca_system_score_gemma":0.002363002,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01033061,"about_ca_topic_score_gemma":0.01620236,"domain_scores_codex":[0.9956191,0.001824723,0.0003648295,0.001172791,0.0007148727,0.0003036855],"domain_scores_gemma":[0.9907634,0.006952446,0.0003348654,0.0007593262,0.0009229835,0.0002670383],"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":[0.001079116,0.000405085,0.001626211,0.0005148179,0.0002952257,0.0002717823,0.0007340495,0.5625248,0.005284357,0.09548573,0.009328437,0.3224505],"study_design_scores_gemma":[0.00001975011,0.00001760695,0.000112683,0.0000176303,0.00001893139,0.00003052508,0.00001931645,0.9746512,0.0004476554,0.02383964,0.0008050704,0.00002007169],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00463613,0.0005837981,0.9934986,0.0001978021,0.00008257759,0.00006842954,0.0002160285,0.000408979,0.0003078033],"genre_scores_gemma":[0.2386214,0.001903603,0.7457706,0.0005809992,0.0008679976,0.001039675,0.003325066,0.0005961182,0.007294542],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01033061,"threshold_uncertainty_score":0.03118229,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04279014910143022,"score_gpt":0.3247153135910042,"score_spread":0.2819251644895739,"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."}}