{"id":"W2970259172","doi":"10.18653/v1/d19-1349","title":"Evaluating Topic Quality with Posterior Variability","year":2019,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Natural language processing; Joint (building); Quality (philosophy); Artificial intelligence; Engineering; Philosophy; Epistemology","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":[0.03258576,0.002307264,0.003192751,0.007379928,0.001551202,0.007033939,0.002421817,0.004778872,0.003412307],"category_scores_gemma":[0.104636,0.001478595,0.002515699,0.004217973,0.001563993,0.007749887,0.004241899,0.0040151,0.002423091],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001545235,"about_ca_system_score_gemma":0.001960736,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007013622,"about_ca_topic_score_gemma":0.005604623,"domain_scores_codex":[0.9863781,0.00688052,0.0009349474,0.003246,0.002017188,0.0005433308],"domain_scores_gemma":[0.9055234,0.07935914,0.002112398,0.007324449,0.004195875,0.001484794],"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":[0.004853325,0.0003498705,0.05162641,0.001635841,0.003131063,0.000380642,0.001186245,0.2878384,0.006811434,0.01558537,0.04998367,0.5766177],"study_design_scores_gemma":[0.0001939241,0.0001783623,0.005797131,0.0001521193,0.0005696795,0.0002395262,0.0001713064,0.9586061,0.002994231,0.02667044,0.004347502,0.0000795739],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1073788,0.01964585,0.8527701,0.002511253,0.000742691,0.0003239058,0.004224529,0.007184208,0.005218611],"genre_scores_gemma":[0.7899425,0.003452097,0.1825381,0.0004719573,0.001644753,0.0002893537,0.01637775,0.002052193,0.00323128],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03258576,"threshold_uncertainty_score":0.172332,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08068371065089569,"score_gpt":0.3653743833317343,"score_spread":0.2846906726808386,"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."}}