{"id":"W3095642204","doi":"10.2196/19735","title":"Measurement of Semantic Textual Similarity in Clinical Texts: Comparison of Transformer-Based Models","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":59,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Center for Chronic Disease Prevention and Health Promotion; Clinical and Translational Science Institute, University of Florida; National Center for Advancing Translational Sciences; University of Florida; National Cancer Institute; National Institutes of Health; Nvidia; Centers for Disease Control and Prevention; National Institute on Aging; Patient-Centered Outcomes Research Institute","keywords":"Computer science; Natural language processing; Semantic similarity; Artificial intelligence; Similarity (geometry); Transformer; Domain (mathematical analysis); Information retrieval; Biomedical text mining; Text mining","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":[0.001624446,0.000140567,0.0006103638,0.00008773701,0.00002003631,0.00001579116,0.001011911,0.0002122251,0.0000182055],"category_scores_gemma":[0.0003105922,0.0001202469,0.000140354,0.0003635544,0.0001416043,0.0003749658,0.0001077363,0.0005223372,0.00000552935],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000344775,"about_ca_system_score_gemma":0.0005293189,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001587062,"about_ca_topic_score_gemma":0.00002422411,"domain_scores_codex":[0.9952531,0.00009468236,0.002392201,0.0001441792,0.001860929,0.0002548552],"domain_scores_gemma":[0.9986591,0.0001500323,0.0003526614,0.0003715163,0.0001599824,0.0003066664],"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":[0.0002776814,0.004563329,0.03542615,0.007827532,0.0001942063,0.00002532855,0.08817185,0.1023495,0.0001895118,0.04525971,0.002765027,0.7129502],"study_design_scores_gemma":[0.001231736,0.0002413047,0.0004635184,0.0002593036,0.000007434105,6.484916e-7,0.0004965989,0.9960827,0.0006032339,0.000335406,0.0001636221,0.0001144439],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1009439,0.00004851922,0.8959957,0.001738865,0.00007124004,0.0003050428,0.000002802584,0.00004527771,0.0008486378],"genre_scores_gemma":[0.9700171,0.000008396303,0.0287712,0.001159656,0.00002771671,0.000009054248,0.00000185641,0.000004541694,4.23592e-7],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8937333,"threshold_uncertainty_score":0.4903527,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1322531638745227,"score_gpt":0.3662089841932468,"score_spread":0.2339558203187241,"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."}}