{"id":"W2979474259","doi":"10.22148/001c.13147","title":"NovelTM Datasets for English-Language Fiction, 1700-2009","year":2020,"lang":"en","type":"article","venue":"Journal of Cultural Analytics","topic":"Digital Humanities and Scholarship","field":"Arts and Humanities","cited_by":37,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"English language; Computer science; Non-fiction; Resilience (materials science); Fragility; History; Linguistics; Literature; Art; Philosophy","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"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":[0.001082663,0.001854397,0.001063992,0.007544381,0.001751915,0.003404704,0.002586758,0.002631814,0.03489392],"category_scores_gemma":[0.006334946,0.0003854427,0.001343588,0.004971207,0.000735118,0.002309595,0.002867523,0.001647198,0.05911956],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001573884,"about_ca_system_score_gemma":0.001673127,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01305211,"about_ca_topic_score_gemma":0.03920003,"domain_scores_codex":[0.9989505,0.0001721737,0.0001450552,0.0002490881,0.0003351029,0.0001481003],"domain_scores_gemma":[0.997799,0.0005188849,0.0001980023,0.0005229645,0.0006991478,0.0002619585],"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":[0.0001250034,0.00008236212,0.001570753,0.001079916,0.00004547193,0.0001955581,0.0001435878,0.0004782292,0.0004629069,0.0008404189,0.9788994,0.01607651],"study_design_scores_gemma":[0.0001155763,0.00005495976,0.007310877,0.0004472501,0.00004741715,0.0003951133,0.0004009719,0.001959451,0.001146839,0.001085632,0.986987,0.00004875675],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.003263062,0.001352997,0.0004398104,0.0003991445,0.0004168937,0.0001017118,0.9877356,0.001400343,0.004890524],"genre_scores_gemma":[0.002547784,0.0002982047,0.001243696,0.00007949785,0.00009991197,0.0001466183,0.9927004,0.00008853593,0.002795345],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.03489392,"threshold_uncertainty_score":0.1167318,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07261985779052788,"score_gpt":0.2707456352499021,"score_spread":0.1981257774593742,"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."}}