{"id":"W3197275000","doi":"10.18653/v1/2021.emnlp-main.181","title":"Unsupervised Conversation Disentanglement through Co-Training","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Conversation; Computer science; Classifier (UML); Session (web analytics); Artificial intelligence; Machine learning; Reinforcement learning; Training set; Speech recognition; World Wide Web","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":[0.002365443,0.002356522,0.001427136,0.001635329,0.0008437277,0.001024576,0.002515421,0.001566844,0.003083415],"category_scores_gemma":[0.00864773,0.0007541566,0.001285241,0.001028,0.0008892327,0.003674195,0.003195519,0.004084158,0.002427146],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000758416,"about_ca_system_score_gemma":0.001531136,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00438437,"about_ca_topic_score_gemma":0.00933098,"domain_scores_codex":[0.9972361,0.0009276456,0.00009624664,0.001221073,0.0002821795,0.0002366774],"domain_scores_gemma":[0.9951775,0.002820145,0.0003651358,0.0006993425,0.0005706384,0.0003671747],"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":[0.001352426,0.0008854615,0.009288321,0.00040053,0.000271831,0.0002978882,0.001433833,0.08755701,0.03732668,0.003880849,0.01036651,0.8469387],"study_design_scores_gemma":[0.00002779297,0.0001135713,0.001378486,0.00002286004,0.00003908344,0.00009037356,0.0001542214,0.9798365,0.009430743,0.005638822,0.00323726,0.00003027279],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06670424,0.001080898,0.9213111,0.0003829105,0.0001565362,0.0002458704,0.0004833981,0.006586153,0.003048972],"genre_scores_gemma":[0.7221346,0.0003271153,0.2630728,0.0004192645,0.0002217147,0.0005777828,0.002810572,0.0006790292,0.009757007],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00438437,"threshold_uncertainty_score":0.01250976,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1138268762368847,"score_gpt":0.4306470469350119,"score_spread":0.3168201706981271,"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."}}