{"id":"W3093553144","doi":"10.18653/v1/2021.naacl-main.139","title":"UmlsBERT: Clinical Domain Knowledge Augmentation of Contextual Embeddings Using the Unified Medical Language System Metathesaurus","year":2021,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Unified Medical Language System; Computer science; Natural language processing; Domain (mathematical analysis); Artificial intelligence; Embedding; Inference; Process (computing); Domain knowledge; Word (group theory); Word embedding; Named-entity recognition; ENCODE; Linguistics; Programming language","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001650225,0.0011775,0.0007470053,0.004437545,0.0006667544,0.001829964,0.001226148,0.001180317,0.008923909],"category_scores_gemma":[0.007946474,0.0005173146,0.001236815,0.002229273,0.0004066729,0.002892421,0.00445679,0.001135722,0.007261337],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006447171,"about_ca_system_score_gemma":0.002196485,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005086549,"about_ca_topic_score_gemma":0.01021334,"domain_scores_codex":[0.9986883,0.0004864027,0.0001482382,0.0003407815,0.0002614459,0.00007491707],"domain_scores_gemma":[0.9983045,0.0006334456,0.0001272312,0.0005025194,0.0003427498,0.00008955759],"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":[0.0006169036,0.0003069142,0.003366656,0.001105289,0.000381861,0.0004584347,0.0007096766,0.007912529,0.00890311,0.008535305,0.15331,0.8143933],"study_design_scores_gemma":[0.0008017356,0.0007190938,0.01179194,0.001463513,0.001032617,0.002178817,0.001740956,0.3278104,0.03457243,0.09758093,0.5200172,0.00029027],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04875692,0.01147507,0.7954692,0.00371233,0.002104587,0.001037581,0.05141476,0.06752836,0.01850103],"genre_scores_gemma":[0.2343737,0.003794806,0.6370239,0.0009798888,0.00056418,0.0009625035,0.1107782,0.002600583,0.008922198],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008923909,"threshold_uncertainty_score":0.02985346,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07782002185679944,"score_gpt":0.3887847771219509,"score_spread":0.3109647552651514,"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."}}