{"id":"W2803982643","doi":"10.2196/medinform.9455","title":"Privacy-Preserving Predictive Modeling: Harmonization of Contextual Embeddings From Different Sources","year":2018,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Machine Learning in Healthcare","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"U.S. National Library of Medicine; National Institute of General Medical Sciences; National Center for Advancing Translational Sciences; National Institute of Biomedical Imaging and Bioengineering; National Human Genome Research Institute","keywords":"Computer science; Pooling; Embedding; Context (archaeology); Word2vec; Data sharing; Data mining; Machine learning; Harmonization; Artificial intelligence; Data science","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":[0.002935944,0.001277095,0.0009143544,0.0007923178,0.0004919347,0.001487452,0.001277235,0.0007657355,0.0007103958],"category_scores_gemma":[0.007873979,0.0005170322,0.001366365,0.001129826,0.001030979,0.003483108,0.003100507,0.00197454,0.0003169759],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006754693,"about_ca_system_score_gemma":0.001073079,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003035355,"about_ca_topic_score_gemma":0.003040239,"domain_scores_codex":[0.9977884,0.0008498745,0.0001287722,0.0006825698,0.0003618174,0.0001884893],"domain_scores_gemma":[0.9972639,0.0009716579,0.0003608254,0.0009594274,0.0003462507,0.00009784347],"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":[0.000317087,0.0002021713,0.008264639,0.00009067686,0.0002743239,0.0002474792,0.0005754685,0.7462093,0.006186324,0.01829035,0.002735395,0.2166068],"study_design_scores_gemma":[0.000007349947,0.00005232479,0.0005171768,0.000009395357,0.00003223025,0.00003917029,0.00006456918,0.9846257,0.002514996,0.01136964,0.0007546341,0.000012873],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04430983,0.0002314201,0.9534019,0.0003995332,0.00003601972,0.00004310946,0.0001985069,0.0005885294,0.0007910897],"genre_scores_gemma":[0.8452305,0.0002843892,0.1517384,0.0002796474,0.00009347891,0.0001149045,0.0009643784,0.0001405964,0.001153557],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003035355,"threshold_uncertainty_score":0.01552695,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02518940963220853,"score_gpt":0.3102796689629046,"score_spread":0.285090259330696,"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."}}