{"id":"W3190643984","doi":"10.5334/johd.42","title":"The Corpus for Idiolectal Research (CIDRE)","year":2021,"lang":"en","type":"article","venue":"Journal of Open Humanities Data","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Agence Nationale de la Recherche","keywords":"Paratext; Scripting language; Metadata; License; Computer science; World Wide Web; Open source; Style (visual arts); Literature; Linguistics; Art; Programming language; Software; Philosophy","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":[0.002936705,0.0007539996,0.001006353,0.01730108,0.003419783,0.003865032,0.001196814,0.001156477,0.04065711],"category_scores_gemma":[0.01092529,0.0003045239,0.0003560031,0.01704996,0.001493519,0.001690161,0.002789537,0.001305154,0.01670372],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005200068,"about_ca_system_score_gemma":0.006072962,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.07552463,"about_ca_topic_score_gemma":0.07976341,"domain_scores_codex":[0.9968388,0.0008986267,0.000359894,0.0007707024,0.0007050975,0.000426799],"domain_scores_gemma":[0.9928323,0.002385525,0.0005535002,0.001424918,0.002291789,0.0005118435],"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":[0.0002298049,0.00008764519,0.005919826,0.002840462,0.00004139943,0.0007831931,0.01153674,0.0002521784,0.003825201,0.05141644,0.7615604,0.1615067],"study_design_scores_gemma":[0.00001840416,0.00000754861,0.008882189,0.0003788226,0.00001071003,0.0002951523,0.001084762,0.00007215051,0.0004420598,0.001297434,0.9874871,0.00002363209],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.04115633,0.009503007,0.007062937,0.001828871,0.0006726086,0.0007479881,0.8390331,0.001661129,0.09833395],"genre_scores_gemma":[0.1266719,0.003167764,0.02703303,0.0007422514,0.0004625967,0.002466179,0.8139523,0.001467204,0.02403672],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.07552463,"threshold_uncertainty_score":0.15017,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.6364291660620014,"score_gpt":0.5070253019086151,"score_spread":0.1294038641533863,"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."}}