{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.005136907,0.0004879246,0.0007146723,0.0009128161,0.0003345215,0.001960789,0.005132278,0.0003189438,0.003786854],"category_scores_gemma":[0.007009666,0.0003828632,0.0003239636,0.001673913,0.0002547429,0.0003158737,0.002048761,0.0006274264,0.01420701],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001983499,"about_ca_system_score_gemma":0.0003164478,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005150954,"about_ca_topic_score_gemma":0.0002276596,"domain_scores_codex":[0.9920106,0.0002235449,0.001215518,0.00206799,0.003792261,0.0006901016],"domain_scores_gemma":[0.9924094,0.0008165946,0.000732968,0.005253924,0.0005057368,0.0002813469],"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":[0.000009947898,0.0001006669,0.00001581035,0.00001808424,0.00002712735,0.00008368036,0.00005110828,0.0001698726,3.509992e-7,0.0001495086,0.9890401,0.01033375],"study_design_scores_gemma":[0.0003935168,0.00005866803,0.0001540055,0.00007282657,0.00004838741,0.000008108344,0.0000828086,0.00004854571,0.000005136029,0.003624091,0.9950557,0.0004481574],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0001129145,0.0002401038,0.0004118099,0.0001829005,0.01398184,0.0003871589,0.9786981,0.0001744197,0.00581069],"genre_scores_gemma":[0.0002309655,0.00004199667,0.0002578863,0.0005624947,0.001229546,0.00002052535,0.986865,0.00002294074,0.01076861],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.0127523,"threshold_uncertainty_score":0.9998623,"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."}}