{"id":"W2401224426","doi":"10.1609/icwsm.v5i1.14198","title":"Supervised Topic Segmentation of Email Conversations","year":2021,"lang":"en","type":"article","venue":"Proceedings of the International AAAI Conference on Web and Social Media","topic":"Personal Information Management and User Behavior","field":"Decision Sciences","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Exploit; Segmentation; Conversation; Artificial intelligence; Natural language processing; Market segmentation; Graph; Information retrieval; Machine learning; Linguistics","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":[0.0003596831,0.00006245042,0.0001285898,0.0001000321,0.00008216798,0.0001152847,0.0003906807,0.00003675082,0.0006346795],"category_scores_gemma":[0.0006089737,0.00004488707,0.00007220845,0.0001858845,0.0001046963,0.0003200473,0.0001246314,0.00006599873,0.000009851927],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001867894,"about_ca_system_score_gemma":0.00006938123,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000004412014,"about_ca_topic_score_gemma":0.00001983959,"domain_scores_codex":[0.9985302,0.000009849382,0.0003425775,0.0001243975,0.0009284577,0.00006452254],"domain_scores_gemma":[0.9983839,0.0001659409,0.0002566952,0.00004366534,0.001127059,0.00002276509],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"observational","study_design_scores_codex":[0.00007441262,0.0001529004,0.09935772,0.00003992152,0.00007905554,5.308659e-7,0.02561222,0.000001556394,0.07896912,0.7499492,0.008201336,0.03756205],"study_design_scores_gemma":[0.003372574,0.00009999055,0.5966169,0.0002070337,0.0001369916,0.000005562556,0.1401735,0.006873046,0.122407,0.1169838,0.01264418,0.0004794073],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9745882,0.000006426221,0.0000101948,0.008487924,0.0003935604,0.00006679372,0.00003598086,0.000006215868,0.0164047],"genre_scores_gemma":[0.9980921,0.00002624067,0.0001631725,0.0002830102,0.00006180177,0.0000064933,0.000007821646,0.000002106648,0.001357309],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6329654,"threshold_uncertainty_score":0.6949295,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2328588741157706,"score_gpt":0.3850110770473913,"score_spread":0.1521522029316208,"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."}}