{"id":"W2980839612","doi":"10.2196/14850","title":"Combining Contextualized Embeddings and Prior Knowledge for Clinical Named Entity Recognition: Evaluation Study","year":2019,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Eli Lilly and Company","keywords":"Computer science; Named-entity recognition; Natural language processing; Artificial intelligence; Word embedding; Lexicon; F1 score; Deep learning; Context (archaeology); Leverage (statistics); Embedding; Information retrieval; Task (project management)","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":[0.004665063,0.0001279592,0.0003363512,0.00007770201,0.0001028839,0.0001481795,0.0004744146,0.0001636645,0.00009852966],"category_scores_gemma":[0.001039863,0.0001110277,0.00006987643,0.0001429712,0.00005843329,0.0006966701,0.0003342725,0.0002804768,0.0001092148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004018859,"about_ca_system_score_gemma":0.0002649168,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002907081,"about_ca_topic_score_gemma":0.000007302233,"domain_scores_codex":[0.9975868,0.0001648704,0.001036044,0.0002049923,0.0007771934,0.0002300855],"domain_scores_gemma":[0.9980311,0.0007499174,0.0002685604,0.0003763643,0.0003477319,0.0002262967],"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":[0.00003416004,0.0004544455,0.004359905,0.0001992177,0.00005744932,8.782656e-7,0.0377051,0.000001761855,0.000001349031,0.001819793,0.001270146,0.9540958],"study_design_scores_gemma":[0.006760767,0.0003703245,0.001783911,0.0001367378,0.00002698922,0.000008285738,0.002383712,0.9831243,0.000005254424,0.001130323,0.004092175,0.000177234],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8627554,0.00005036956,0.1328236,0.0001907295,0.0008873799,0.002040394,0.00000173861,0.0001070798,0.001143312],"genre_scores_gemma":[0.9725004,0.0000164593,0.02641517,0.0006252141,0.0001125395,0.0002206599,0.00001343408,0.000007361136,0.00008879478],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9831225,"threshold_uncertainty_score":0.4527577,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1175884097070168,"score_gpt":0.4348812763784703,"score_spread":0.3172928666714535,"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."}}