{"id":"W3165206559","doi":"10.2196/23099","title":"Predicting Semantic Similarity Between Clinical Sentence Pairs Using Transformer Models: Evaluation and Representational Analysis","year":2021,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":33,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Natural language processing; Semantic similarity; Computer science; Artificial intelligence; Sentence; Security token; Similarity (geometry); Transformer","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":[0.00815073,0.001930338,0.0009649037,0.002384322,0.0006383419,0.001824623,0.002560171,0.001665118,0.002083288],"category_scores_gemma":[0.02673447,0.0005021605,0.00193518,0.001298823,0.0008473538,0.003012882,0.002164083,0.002297698,0.0009437779],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003368515,"about_ca_system_score_gemma":0.002272899,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01743489,"about_ca_topic_score_gemma":0.01421489,"domain_scores_codex":[0.9961523,0.001649649,0.0003503038,0.0009060887,0.0007382353,0.0002032581],"domain_scores_gemma":[0.9809715,0.01476907,0.0006549542,0.001334214,0.001734006,0.0005362711],"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":[0.003808156,0.001393422,0.05539735,0.0007995337,0.001230756,0.0007849937,0.000954453,0.6033831,0.009752919,0.004049203,0.01035418,0.308092],"study_design_scores_gemma":[0.00003487004,0.0002053151,0.001181628,0.0000145832,0.00007250607,0.00008619798,0.00007176428,0.9938688,0.002678896,0.001471031,0.0002982736,0.00001605524],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7833918,0.001197174,0.1936051,0.00145249,0.0002799409,0.0007649966,0.004005732,0.01137172,0.003931001],"genre_scores_gemma":[0.9492649,0.0001924181,0.04433273,0.0001803691,0.00003668432,0.0001982793,0.004917589,0.000182193,0.0006947893],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01743489,"threshold_uncertainty_score":0.04310566,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1501242600776237,"score_gpt":0.413109033859521,"score_spread":0.2629847737818973,"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."}}