{"id":"W6926263748","doi":"10.21415/t5161t","title":"ClassBank English Roth Corpus","year":2004,"lang":"en","type":"dataset","venue":"TalkBank","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Subject (documents); Feature (linguistics); Corpus linguistics","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001677384,0.002788404,0.001648853,0.006886689,0.001828815,0.00360249,0.002545023,0.00237879,0.238348],"category_scores_gemma":[0.00645546,0.001266772,0.0009480878,0.01027484,0.0007861003,0.002400588,0.002472772,0.002530213,0.2103311],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002980599,"about_ca_system_score_gemma":0.005281334,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.06464174,"about_ca_topic_score_gemma":0.0804514,"domain_scores_codex":[0.9984493,0.0003273058,0.0001844723,0.0004881658,0.0003487781,0.0002020076],"domain_scores_gemma":[0.9965093,0.001135758,0.0001901525,0.0008761497,0.001006718,0.0002819136],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00005150116,0.00001793367,0.0001109232,0.0001687116,0.000008063192,0.0000156277,0.00002820182,0.0001146221,0.0001573216,0.0007120172,0.9963464,0.00226869],"study_design_scores_gemma":[0.0001545462,0.0000113554,0.002562536,0.0001266098,0.00002514082,0.00004538958,0.0001352206,0.0003975724,0.0007292336,0.001212587,0.9945722,0.00002758042],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0004121511,0.0001058656,0.0002822737,0.0001503799,0.00009272232,0.00002387272,0.9924874,0.0007159639,0.005729362],"genre_scores_gemma":[0.001030772,0.00006699663,0.0004749369,0.00004914333,0.00001616671,0.0001340361,0.992826,0.0004438388,0.004958168],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.238348,"threshold_uncertainty_score":0.7973534,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1021480648467456,"score_gpt":0.3757745092181958,"score_spread":0.2736264443714502,"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."}}