{"id":"W2990235080","doi":"10.33011/computel.v1i.345","title":"OCR Evaluation Tools for the 21st Century","year":2019,"lang":"en","type":"article","venue":"","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"","keywords":"Unicode; Computer science; Natural language processing; Word (group theory); Character (mathematics); Confusion; Artificial intelligence; Speech recognition; Linguistics; Psychology","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01256686,0.003088201,0.001459179,0.01132321,0.00145068,0.004908314,0.00346178,0.001780601,0.04745802],"category_scores_gemma":[0.06235943,0.001056515,0.001318294,0.004475493,0.001141734,0.006115099,0.003744369,0.002046374,0.03188205],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001933295,"about_ca_system_score_gemma":0.002372589,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003767887,"about_ca_topic_score_gemma":0.003518279,"domain_scores_codex":[0.9824258,0.002883196,0.002922576,0.0015417,0.009738061,0.0004886435],"domain_scores_gemma":[0.9356369,0.01386029,0.004175465,0.01038882,0.03482881,0.001109625],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0003452617,0.0001417438,0.002836345,0.001363003,0.0001284918,0.0003806694,0.0006810207,0.003278958,0.02435289,0.01950113,0.2058876,0.7411029],"study_design_scores_gemma":[0.0001497824,0.000350826,0.008778914,0.001075715,0.0001478538,0.00329544,0.0004907735,0.0462386,0.121452,0.02642274,0.7910048,0.0005925298],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005329519,0.001908452,0.7035728,0.0007889043,0.000634241,0.0008096608,0.008673661,0.2502682,0.02801461],"genre_scores_gemma":[0.04137687,0.001178075,0.8580855,0.0007194561,0.0003985565,0.001453763,0.02500913,0.04974677,0.02203186],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.04745802,"threshold_uncertainty_score":0.1587629,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04110869910776668,"score_gpt":0.2936906192732608,"score_spread":0.2525819201654942,"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."}}