{"id":"W4389941462","doi":"10.1007/s10032-023-00456-5","title":"Improving accuracy and explainability of online handwritten character recognition","year":2023,"lang":"en","type":"article","venue":"International Journal on Document Analysis and Recognition (IJDAR)","topic":"Time Series Analysis and Forecasting","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo; National Research Council Canada","funders":"","keywords":"Computer science; Artificial intelligence; Handwriting; Robustness (evolution); Pattern recognition (psychology); Handwriting recognition; Machine learning; Task (project management); Deep learning; Alphabet; Speech recognition; Feature extraction","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009404945,0.0001686115,0.0003326251,0.001076827,0.000194032,0.0005577914,0.000318891,0.00006367111,0.0002215584],"category_scores_gemma":[0.0002614808,0.0001365856,0.0002801428,0.0008322989,0.00005639981,0.001068405,0.0002080416,0.0002202092,0.00001936639],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005533707,"about_ca_system_score_gemma":0.00003049101,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007575301,"about_ca_topic_score_gemma":0.00003472724,"domain_scores_codex":[0.9980459,0.0001109482,0.0006870687,0.0003727351,0.0005802111,0.0002031145],"domain_scores_gemma":[0.9982071,0.0002299262,0.0006424779,0.0001608505,0.0006252703,0.0001343884],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.00007042774,0.000128318,0.006037605,0.0000158017,0.001252978,0.00003907484,0.0004060748,0.00004927519,0.0009504187,0.0001978371,0.00007994211,0.9907722],"study_design_scores_gemma":[0.006605598,0.001491957,0.5405193,0.0009560467,0.003509109,0.0005773957,0.003109926,0.3380943,0.01328843,0.08480094,0.004794212,0.002252825],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9793159,0.000061997,0.01743888,0.002569107,0.0003055721,0.00008539737,0.00005607918,0.00004593639,0.000121081],"genre_scores_gemma":[0.9945189,0.001047034,0.003533585,0.0002461283,0.000280448,0.00000686046,0.0002138175,0.000008352406,0.0001448828],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9885194,"threshold_uncertainty_score":0.5569798,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02193815844080075,"score_gpt":0.2769692219164192,"score_spread":0.2550310634756184,"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."}}