{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007285238,0.002279032,0.001539621,0.001986274,0.0005132755,0.00114601,0.00142348,0.002083156,0.002529288],"category_scores_gemma":[0.01684155,0.0003570634,0.001321684,0.001269957,0.000652903,0.0028517,0.001737549,0.001063901,0.001446188],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001081386,"about_ca_system_score_gemma":0.001192451,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006081423,"about_ca_topic_score_gemma":0.007140953,"domain_scores_codex":[0.9959596,0.00163239,0.0005587628,0.0008575482,0.0007888153,0.00020291],"domain_scores_gemma":[0.9896575,0.006090244,0.0005528278,0.001168305,0.00199079,0.0005404201],"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":[0.008589764,0.004758471,0.06948738,0.003878941,0.00279601,0.000932178,0.0003466047,0.05973757,0.009697278,0.0007291046,0.03437424,0.8046725],"study_design_scores_gemma":[0.001347726,0.01021684,0.06147161,0.0005439025,0.002959608,0.002415714,0.0006369425,0.8827564,0.02232365,0.002106104,0.01295807,0.0002634162],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9239374,0.02208798,0.03132309,0.001178936,0.001077869,0.001274699,0.007050811,0.00401909,0.00805011],"genre_scores_gemma":[0.9290953,0.00459493,0.03867093,0.0004609654,0.000365943,0.0004750418,0.02389575,0.0001464781,0.002294736],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007285238,"threshold_uncertainty_score":0.0385285,"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."}}