{"id":"W3109919947","doi":"10.2196/22508","title":"Identification of Semantically Similar Sentences in Clinical Notes: Iterative Intermediate Training Using Multi-Task Learning","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Task (project management); Identification (biology); Natural language processing; Training (meteorology); Artificial intelligence; Engineering","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":[0.00151584,0.000141595,0.000385818,0.0001099623,0.00005233588,0.00009152827,0.0007911095,0.0001743689,0.00001138019],"category_scores_gemma":[0.002669945,0.0001248667,0.00009423168,0.0003967064,0.000173662,0.0007898508,0.0003475586,0.0007051995,0.00001624396],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002583864,"about_ca_system_score_gemma":0.0002287854,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000405339,"about_ca_topic_score_gemma":0.000003644985,"domain_scores_codex":[0.9965794,0.0001784602,0.002039364,0.0002022281,0.0007244954,0.0002760359],"domain_scores_gemma":[0.9983848,0.0004814383,0.0005234014,0.000222451,0.0001161058,0.0002717759],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004708176,0.0004047967,0.04963893,0.001046711,0.0001089284,0.0001094924,0.5898356,0.01759993,0.001344856,0.003788121,0.00003945781,0.3360361],"study_design_scores_gemma":[0.0005492553,0.00007332904,0.001767982,0.0002389907,0.000004830232,0.000005633552,0.003826177,0.9931428,0.0001350511,0.0000809225,0.00005192462,0.0001231327],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3971055,0.000008734216,0.6018596,0.0006858355,0.0001397272,0.0001305432,8.69542e-7,0.0000462511,0.00002290559],"genre_scores_gemma":[0.9310282,0.00001989634,0.06768782,0.001156686,0.00008907012,0.000006729158,0.000003211604,0.000006033786,0.000002313969],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9755428,"threshold_uncertainty_score":0.5091915,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1071406895850211,"score_gpt":0.3789297631923781,"score_spread":0.271789073607357,"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."}}