{"id":"W2057330140","doi":"10.1145/1753326.1753328","title":"Estimating residual error rate in recognized handwritten documents using artificial error injection","year":2010,"lang":"en","type":"article","venue":"","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Residual; Handwriting; Word error rate; Handwriting recognition; Set (abstract data type); Speech recognition; Measure (data warehouse); Artificial intelligence; Pattern recognition (psychology); Data mining; Feature extraction; Algorithm","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":[0.001507644,0.0002366705,0.0002801045,0.000497227,0.0002267457,0.0004383365,0.0005726928,0.000201728,0.0002196265],"category_scores_gemma":[0.0003936149,0.0002324207,0.00006896536,0.000769971,0.00006763605,0.001469191,0.0002622977,0.0005485984,0.0001178512],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007542442,"about_ca_system_score_gemma":0.0001310485,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005745314,"about_ca_topic_score_gemma":0.0008929134,"domain_scores_codex":[0.9977287,0.0002136535,0.0006374761,0.0006357236,0.0003084907,0.0004759218],"domain_scores_gemma":[0.998899,0.0001209326,0.0002041467,0.000479869,0.0001752962,0.0001207181],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001086836,0.0005579184,0.003943842,0.00006981821,0.00004946388,0.0001255139,0.001566217,0.0003649844,0.4503975,0.009177036,0.001323916,0.5323151],"study_design_scores_gemma":[0.001162297,0.0002061358,0.003015408,0.0001483462,0.00001828342,0.0001660247,0.0001038858,0.6687248,0.2388279,0.08663747,0.0001755693,0.0008138944],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5821372,0.00000220042,0.4141866,0.0004183252,0.0008476191,0.0003799018,0.000001943297,0.0005837847,0.001442491],"genre_scores_gemma":[0.5225544,3.626188e-7,0.4767997,0.0002357457,0.0001753041,0.00004553672,0.000005739598,0.00001788549,0.0001653634],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6683598,"threshold_uncertainty_score":0.9477842,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03983830630952655,"score_gpt":0.3224563889916165,"score_spread":0.28261808268209,"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."}}