{"id":"W2250174871","doi":"","title":"Classifying English Documents by National Dialect","year":2013,"lang":"en","type":"article","venue":"","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":42,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"British National Corpus; Linguistics; Variety (cybernetics); Variation (astronomy); Computer science; Natural language processing; Intuition; Artificial intelligence; Psychology","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":[0.001458323,0.0004795828,0.0005140118,0.009016614,0.001003089,0.002346317,0.0003741366,0.000420092,0.002317403],"category_scores_gemma":[0.009384949,0.0001542377,0.0003753477,0.007450703,0.000448046,0.001583727,0.0009061313,0.0004206231,0.001324794],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001059711,"about_ca_system_score_gemma":0.001141982,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02838944,"about_ca_topic_score_gemma":0.05389735,"domain_scores_codex":[0.9988298,0.0002227694,0.000208686,0.0002739732,0.0003136721,0.0001511622],"domain_scores_gemma":[0.9906883,0.00442545,0.0008761577,0.000863582,0.00282074,0.0003258167],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005742231,0.0003762207,0.4971851,0.00153329,0.0003176052,0.0006198373,0.005889709,0.002906125,0.03021995,0.003983611,0.02769412,0.4287003],"study_design_scores_gemma":[0.00006755948,0.0001807093,0.7812376,0.0003810341,0.0003823051,0.001347601,0.01077878,0.04765545,0.03640736,0.004752112,0.116669,0.0001405085],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.9383478,0.003576119,0.01245818,0.0005826675,0.0001328736,0.000263947,0.0269882,0.0009031879,0.01674714],"genre_scores_gemma":[0.914391,0.001611522,0.03912862,0.0001163819,0.000101218,0.000175538,0.0373931,0.0001306404,0.006951964],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.02838944,"threshold_uncertainty_score":0.0564484,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02166335923556349,"score_gpt":0.2665022127154754,"score_spread":0.2448388534799119,"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."}}