{"id":"W2898607795","doi":"10.2196/medinform.9965","title":"Clinical Named Entity Recognition From Chinese Electronic Health Records via Machine Learning Methods","year":2018,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":57,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Conditional random field; Artificial intelligence; Natural language processing; F1 score; Machine learning; Named-entity recognition; Test set; Benchmark (surveying); Precision and recall; Health records; Deep learning; Information retrieval; Unstructured data; Task (project management); Data mining; Big data; Health care","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004748725,0.0002001009,0.0004467317,0.0001100716,0.0002212844,0.0001045397,0.0009754004,0.0002640624,0.000426121],"category_scores_gemma":[0.001129019,0.000162506,0.0001300729,0.0004092241,0.0001382689,0.0007126875,0.0004804233,0.001334041,0.000332184],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001280465,"about_ca_system_score_gemma":0.0005144926,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003002449,"about_ca_topic_score_gemma":0.0002286634,"domain_scores_codex":[0.9962907,0.0006461648,0.001437914,0.000262483,0.0007777843,0.0005849334],"domain_scores_gemma":[0.9977372,0.0005438772,0.0005002217,0.0005794495,0.0001369547,0.0005023439],"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":[0.000009627071,0.00006550473,0.002207399,0.00002959348,0.00002552671,0.000001289441,0.002698948,0.000001619301,0.000002640051,0.0001956722,0.0004155341,0.9943466],"study_design_scores_gemma":[0.0006997795,0.0003959317,0.001079264,0.00007690755,0.000005060494,0.0000181286,0.00005158444,0.9664919,0.00002031784,0.009247073,0.02171324,0.0002008124],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.132144,0.0001033717,0.864374,0.001102716,0.001058115,0.0002173727,0.000002692906,0.0003004936,0.00069726],"genre_scores_gemma":[0.2072625,0.0003166686,0.7852067,0.005864664,0.001123771,0.00003261961,0.00008782296,0.00001744424,0.00008772191],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9941458,"threshold_uncertainty_score":0.66268,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04074273956746675,"score_gpt":0.4129983582792558,"score_spread":0.372255618711789,"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."}}