{"id":"W3111991384","doi":"10.2196/23086","title":"ALBERT-Based Self-Ensemble Model With Semisupervised Learning and Data Augmentation for Clinical Semantic Textual Similarity Calculation: Algorithm Validation Study","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Natural Science Foundation of China","keywords":"Computer science; Semantic similarity; Similarity (geometry); Artificial intelligence; Process (computing); Natural language processing; Set (abstract data type); GRASP; Pearson product-moment correlation coefficient; Data set; Machine learning; Information retrieval; Task (project management); Statistics; Mathematics","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":[0.001736245,0.0002006993,0.0003373479,0.00005956571,0.0002786313,0.000205809,0.0007809162,0.0001779209,0.000009937919],"category_scores_gemma":[0.0007400061,0.0001682076,0.00003605683,0.0003164373,0.00005713933,0.0009092288,0.0004599026,0.0006746189,0.000007669865],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000361588,"about_ca_system_score_gemma":0.0004887894,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001818704,"about_ca_topic_score_gemma":0.0000109539,"domain_scores_codex":[0.9970238,0.000279389,0.0009564779,0.0003889525,0.001066328,0.0002850677],"domain_scores_gemma":[0.9977067,0.0007799833,0.0003214823,0.0005183154,0.0002031432,0.000470409],"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":[0.000284223,0.001562794,0.1928187,0.00330074,0.000320757,0.00003905322,0.06363492,0.05408779,0.000004083489,0.0009384519,0.006384965,0.6766235],"study_design_scores_gemma":[0.00261601,0.0008459873,0.001361821,0.00005391863,0.00003938621,0.000006926854,0.0008989613,0.9932277,0.000005125845,0.00002135408,0.000720925,0.0002018635],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05841898,0.000009831996,0.9354311,0.004466172,0.00005127997,0.00125532,0.000009669376,0.0002950642,0.00006257034],"genre_scores_gemma":[0.5305856,0.00001169498,0.4659594,0.002780711,0.0001452876,0.00008304323,0.0003994288,0.00001902072,0.00001580665],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9391399,"threshold_uncertainty_score":0.6859307,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09368095229211959,"score_gpt":0.4181064391805727,"score_spread":0.3244254868884531,"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."}}