{"id":"W3202078810","doi":"10.2196/26407","title":"Chinese-Named Entity Recognition From Adverse Drug Event Records: Radical Embedding-Combined Dynamic Embedding–Based BERT in a Bidirectional Long Short-term Conditional Random Field (Bi-LSTM-CRF) Model","year":2021,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Pharmacovigilance and Adverse Drug Reactions","field":"Pharmacology, Toxicology and Pharmaceutics","cited_by":25,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"China Pharmaceutical University; National Natural Science Foundation of China","keywords":"Conditional random field; Pharmacovigilance; Named-entity recognition; Computer science; Artificial intelligence; Natural language processing; Drug reaction; Recall; Adverse drug reaction; Medicine; Machine learning; Adverse effect; Drug; Pharmacology; Engineering; Psychology; Cognitive psychology","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":["metaepi_narrow","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0006925114,0.0004262096,0.0006356054,0.0003087185,0.0003290514,0.00004686355,0.0003511832,0.0006585672,0.008198421],"category_scores_gemma":[0.000591132,0.0004197391,0.0003725744,0.0005254438,0.0002562593,0.000759431,0.0001418618,0.002287388,0.0002353206],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003724278,"about_ca_system_score_gemma":0.0008885282,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003427108,"about_ca_topic_score_gemma":0.0002755675,"domain_scores_codex":[0.9963676,0.0003784568,0.001342938,0.0003719004,0.0008965922,0.0006424914],"domain_scores_gemma":[0.9970571,0.001517843,0.0002848469,0.000260899,0.0001976203,0.0006816563],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.02128188,0.02812206,0.3335468,0.002875082,0.005335619,0.005186184,0.0239835,0.2487996,0.01516308,0.0004099807,0.1948919,0.1204042],"study_design_scores_gemma":[0.01066623,0.00004542032,0.00384395,0.0002438662,0.0001641032,0.00006726247,0.0003963993,0.9775383,0.002129689,0.001226013,0.00317415,0.0005045534],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9814868,0.00008198881,0.01152603,0.001714257,0.002109733,0.0006998905,0.0006931424,0.0002016113,0.001486537],"genre_scores_gemma":[0.9848318,0.0004683648,0.00123101,0.007408915,0.0003549588,0.0004152827,0.00483198,0.00003199914,0.0004257148],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7287387,"threshold_uncertainty_score":0.9998254,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03818304628788474,"score_gpt":0.4215045395778254,"score_spread":0.3833214932899406,"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."}}