{"id":"W1685527554","doi":"10.1016/j.jbi.2015.09.004","title":"Hidden Markov model using Dirichlet process for de-identification","year":2015,"lang":"en","type":"article","venue":"Journal of Biomedical Informatics","topic":"Bayesian Methods and Mixture Models","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":true,"ca_institutions":"Memorial University of Newfoundland","funders":"U.S. National Library of Medicine","keywords":"Hidden Markov model; Computer science; Conditional random field; Artificial intelligence; Dirichlet process; Feature (linguistics); Context (archaeology); Identification (biology); Machine learning; Dirichlet distribution; Hierarchical Dirichlet process; Vocabulary; Pattern recognition (psychology); Field (mathematics); Bayesian probability; Topic model; Latent Dirichlet allocation; Mathematics","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":[0.002582972,0.0000869713,0.0002014308,0.0001912535,0.00005089731,0.0001287927,0.0006679522,0.0001042663,8.317774e-7],"category_scores_gemma":[0.0003697692,0.00006445082,0.00008248624,0.0002700991,0.0000524736,0.001006956,0.00006465783,0.0001598295,0.000001338866],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009699764,"about_ca_system_score_gemma":0.0006764988,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":5.92391e-7,"about_ca_topic_score_gemma":6.783745e-8,"domain_scores_codex":[0.9983453,0.00003486126,0.0008344403,0.00005117133,0.0005257851,0.0002084464],"domain_scores_gemma":[0.9982479,0.00006789406,0.0006450941,0.0001767824,0.0005128287,0.0003494921],"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":[0.00009597219,0.0003611996,0.00004774235,0.0005718357,0.0001247931,0.00001575286,0.02651125,0.003013218,0.001620247,0.01295229,0.03300073,0.921685],"study_design_scores_gemma":[0.0004518561,0.00008775407,0.000005303806,0.00004294512,0.00002017194,0.000166858,0.0001276724,0.9269866,0.0004817395,0.07080599,0.0007444712,0.00007869244],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01063294,0.00005450369,0.9878895,0.0008114631,0.0003410743,0.00009778165,0.000003490798,0.00001437904,0.0001548704],"genre_scores_gemma":[0.05794127,0.000009204176,0.941393,0.0004542266,0.000156224,0.000002086783,0.000001393934,0.000005233664,0.000037348],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9239733,"threshold_uncertainty_score":0.2628228,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06895016093744245,"score_gpt":0.36020037196107,"score_spread":0.2912502110236276,"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."}}