{"id":"W3022739968","doi":"10.1007/978-3-030-47358-7_51","title":"Improving Classification Using Topic Correlation and Expectation Propagation","year":2020,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Concordia University","funders":"","keywords":"Latent Dirichlet allocation; Computer science; Dirichlet distribution; Topic model; Probabilistic logic; Artificial intelligence; Inference; Marginal distribution; Independence (probability theory); Machine learning; Mathematics; Statistics; Random variable","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007478838,0.0001564641,0.00019604,0.0003319036,0.0004841566,0.0002380648,0.0002197617,0.0001610344,0.00002465459],"category_scores_gemma":[0.0003114386,0.0001590787,0.00004643127,0.0004001329,0.0004198521,0.0003659341,0.00008895616,0.0002509799,0.000004707829],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003113564,"about_ca_system_score_gemma":0.0004307885,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002295373,"about_ca_topic_score_gemma":0.0002344862,"domain_scores_codex":[0.9982896,0.00008712348,0.0002877393,0.0005687414,0.0005960714,0.0001707455],"domain_scores_gemma":[0.9989418,0.0003358332,0.0002882173,0.0001300827,0.0002210347,0.00008305925],"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":[0.000004250546,0.000005574832,0.0004670051,0.00001884527,0.000005053283,0.000002554676,0.003943464,0.01729022,0.0006008219,0.03389212,7.406816e-7,0.9437693],"study_design_scores_gemma":[0.00007756342,0.00002382756,0.00131164,0.0000617614,0.00002027403,0.000001741058,0.000007387865,0.9222462,0.00006142865,0.07582924,0.0001692668,0.0001896644],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.001693761,0.0001256359,0.9943499,0.0006833202,0.0004524558,0.0002736974,8.428622e-7,0.00004778426,0.002372576],"genre_scores_gemma":[0.8060125,0.00001939787,0.1928278,0.0001896679,0.0007343177,0.000003729731,0.00001360771,0.00001297848,0.0001859802],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9435797,"threshold_uncertainty_score":0.6487041,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05629440975015702,"score_gpt":0.3279848024457549,"score_spread":0.2716903926955979,"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."}}