{"id":"W4385144683","doi":"10.1186/s12911-023-02216-1","title":"Quality indices for topic model selection and evaluation: a literature review and case study","year":2023,"lang":"en","type":"review","venue":"BMC Medical Informatics and Decision Making","topic":"Topic Modeling","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"Public Health Ontario; University of Toronto","funders":"Canadian Institutes of Health Research","keywords":"Computer science; Topic model; Artificial intelligence; Automatic summarization; Natural language processing; Model selection; Statistic; Information retrieval; Cluster analysis; Machine learning; Data mining; Statistics; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005900632,0.0002395456,0.0009641041,0.0002089872,0.000221182,0.0004726106,0.0002784496,0.0002454553,0.000002790862],"category_scores_gemma":[0.001167306,0.0001659403,0.0000839101,0.0004044653,0.00002010313,0.0004022463,0.0005407098,0.0003151135,9.665162e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003170393,"about_ca_system_score_gemma":0.0003722092,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002631705,"about_ca_topic_score_gemma":0.00005722658,"domain_scores_codex":[0.9971426,0.0001730923,0.001336212,0.0003206202,0.0008511498,0.0001763963],"domain_scores_gemma":[0.9968279,0.002003934,0.000491447,0.000343166,0.000174997,0.0001585485],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[7.375663e-7,0.000008713982,0.000003899864,0.03557449,0.00001763152,0.00001077425,0.0006563351,0.00002290199,1.260906e-10,0.0006207417,0.00005936427,0.9630244],"study_design_scores_gemma":[0.0003078665,0.00004281495,0.00000110886,0.03963134,0.0002411049,0.001167473,0.0001396313,0.9498578,6.285815e-10,0.002871332,0.005558338,0.0001811424],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.0001621726,0.5239989,0.4747304,0.000005184267,0.00006606567,0.0009978509,0.00000326683,0.00002498171,0.00001121288],"genre_scores_gemma":[0.00003775293,0.7981548,0.2013832,0.000193373,0.00005396326,0.0001583339,0.000004546947,0.000007516137,0.000006443866],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9628432,"threshold_uncertainty_score":0.676685,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2952128063180214,"score_gpt":0.5148011429221888,"score_spread":0.2195883366041674,"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."}}