{"id":"W3033682799","doi":"10.2196/18910","title":"Reliability of Supervised Machine Learning Using Synthetic Data in Health Care: Model to Preserve Privacy for Data Sharing","year":2020,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":164,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Horizon 2020 Framework Programme; European Commission","keywords":"Synthetic data; Machine learning; Computer science; Artificial intelligence; Random forest; Decision tree; Support vector machine; Generalizability theory; Data mining; 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":["metaresearch","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.002808229,0.0002100934,0.0005100409,0.0001481006,0.0001030797,0.0001189527,0.07274007,0.0001798532,0.000009563107],"category_scores_gemma":[0.09358283,0.0001927567,0.00003481019,0.0007789507,0.0001076456,0.002608313,0.2674657,0.0007268533,0.000003821685],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001408259,"about_ca_system_score_gemma":0.0006516877,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001984152,"about_ca_topic_score_gemma":0.00002646544,"domain_scores_codex":[0.9962461,0.00007820796,0.001449096,0.0006770248,0.0009951548,0.0005544378],"domain_scores_gemma":[0.981392,0.0004172313,0.0003571034,0.01731072,0.0001201952,0.0004027874],"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":[0.000670567,0.001562608,0.02320031,0.05558467,0.0002620775,0.00004801121,0.1471804,0.08986101,0.000450938,0.005492823,0.2321336,0.443553],"study_design_scores_gemma":[0.0005300288,0.0001128107,0.00003047314,0.0003379884,0.000004349483,0.000002879328,0.0003029119,0.99463,0.00006130926,0.0018955,0.001911305,0.0001804856],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03779091,0.0001249076,0.9346687,0.02539369,0.00006927026,0.001122745,0.0004354553,0.0003551312,0.00003916757],"genre_scores_gemma":[0.2050835,0.00003978672,0.7931794,0.001131074,0.00002127414,0.00002517953,0.0005027241,0.00001619077,8.680049e-7],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9047689,"threshold_uncertainty_score":0.9322768,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.152395735813922,"score_gpt":0.379612199091731,"score_spread":0.227216463277809,"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."}}