{"id":"W3136599387","doi":"10.1109/access.2021.3096823","title":"uTHCD: A New Benchmarking for Tamil Handwritten OCR","year":2021,"lang":"en","type":"preprint","venue":"IEEE Access","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Trent University; Nottingham Trent University","keywords":"Computer science; Tamil; Database; Grid; Benchmark (surveying); Benchmarking; Font; Optical character recognition; Artificial intelligence; Scripting language; Metadata; Field (mathematics); Information retrieval; Natural language processing; Pattern recognition (psychology); World Wide Web; Image (mathematics)","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"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":[0.002435106,0.002097857,0.0009532243,0.006227395,0.0008276713,0.001649653,0.003195008,0.001304938,0.006123736],"category_scores_gemma":[0.006529947,0.0003273661,0.0006608359,0.004246198,0.0005796294,0.002011609,0.001787092,0.0007163464,0.00486656],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001077311,"about_ca_system_score_gemma":0.0007486227,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005577423,"about_ca_topic_score_gemma":0.005934604,"domain_scores_codex":[0.9959069,0.0005255004,0.0005449025,0.0009549892,0.001759086,0.0003084938],"domain_scores_gemma":[0.9948493,0.0006245609,0.0004258834,0.00182637,0.001894197,0.0003797089],"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":[0.002088897,0.001272075,0.0240509,0.002663119,0.0004606315,0.001034845,0.0003873952,0.03028066,0.06689744,0.003380538,0.1740948,0.6933888],"study_design_scores_gemma":[0.0004600894,0.003941906,0.09842786,0.000553716,0.0002980549,0.00543943,0.000957011,0.3092306,0.3009415,0.002890042,0.2765122,0.0003476747],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5561833,0.00809116,0.1487853,0.0006281835,0.00175313,0.002083639,0.1318662,0.09994643,0.05066267],"genre_scores_gemma":[0.4976719,0.001163576,0.1346286,0.0003069896,0.000181134,0.001004322,0.349779,0.003530719,0.0117338],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006227395,"threshold_uncertainty_score":0.02048588,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06194948546271989,"score_gpt":0.3390967646218885,"score_spread":0.2771472791591686,"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."}}