{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003869927,0.000290704,0.0007625837,0.0001070538,0.0001036113,0.00022061,0.002102708,0.0004663501,0.00007581636],"category_scores_gemma":[0.0003296813,0.0002030993,0.0003889508,0.0002368257,0.0001527622,0.0001540017,0.002853904,0.0008499289,0.000009568651],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001566286,"about_ca_system_score_gemma":0.000971193,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006950802,"about_ca_topic_score_gemma":0.000208063,"domain_scores_codex":[0.9955514,0.001087851,0.001368841,0.0008359686,0.000858777,0.0002971721],"domain_scores_gemma":[0.9967071,0.0008786914,0.0005461312,0.001437842,0.0002589238,0.0001713432],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007476697,0.0008422847,0.000739112,0.002739629,0.001785164,0.0006141597,0.1125845,0.004611915,0.002157118,0.7352192,0.0005521537,0.13808],"study_design_scores_gemma":[0.0008071929,0.00003067998,0.0001044158,0.0008573935,0.0000888822,0.00007848151,0.01440345,0.9819131,0.0008260476,0.0004288087,0.0001242294,0.0003373331],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.28137,0.0007410501,0.7132261,0.0004003944,0.001648011,0.0003156381,0.000002923602,0.0001317222,0.002164079],"genre_scores_gemma":[0.8768665,0.000018846,0.1223183,0.0002069332,0.000307323,0.00001857369,0.00001142218,0.00001880524,0.0002333062],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9773012,"threshold_uncertainty_score":0.8282148,"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."}}