{"id":"W3103720397","doi":"10.2196/23357","title":"Using Character-Level and Entity-Level Representations to Enhance Bidirectional Encoder Representation From Transformers-Based Clinical Semantic Textual Similarity Model: ClinicalSTS Modeling Study","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Harbin Institute of Technology","keywords":"Computer science; Natural language processing; Snippet; Artificial intelligence; Information retrieval; Encoder; Semantic similarity; Transformer; Sentence; Language model","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":[0.0008278855,0.0006913557,0.000401776,0.001110223,0.0002291573,0.0006984627,0.001287545,0.0008489629,0.002307901],"category_scores_gemma":[0.004327206,0.0002308159,0.001031804,0.0008830194,0.0003877631,0.001685282,0.0006864265,0.001091191,0.0005744732],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00135608,"about_ca_system_score_gemma":0.001273499,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01279096,"about_ca_topic_score_gemma":0.01375591,"domain_scores_codex":[0.9994739,0.0001577728,0.00003673732,0.0001718249,0.0001044051,0.00005553325],"domain_scores_gemma":[0.9983865,0.0009902347,0.0001521681,0.0001633756,0.0002450797,0.0000625708],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005300832,0.0004580602,0.00980642,0.0001830609,0.0001456216,0.0004380308,0.0002955546,0.7070616,0.009515078,0.02010962,0.004349442,0.2471074],"study_design_scores_gemma":[0.000005873495,0.00002893308,0.0003153973,0.000003841649,0.00001212269,0.00003235775,0.00001050532,0.995948,0.001155649,0.00211574,0.0003670222,0.000004491981],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2136276,0.0006083919,0.7765166,0.0009148425,0.00007592985,0.0002476154,0.001478331,0.002450386,0.004080373],"genre_scores_gemma":[0.8666537,0.0003148555,0.1248021,0.0001952626,0.0000480592,0.0002200895,0.00275056,0.000126189,0.004889267],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01279096,"threshold_uncertainty_score":0.02543306,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2662994500200416,"score_gpt":0.4595541498079628,"score_spread":0.1932546997879213,"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."}}