{"id":"W3134285077","doi":"10.2196/23728","title":"Learning From Others Without Sacrificing Privacy: Simulation Comparing Centralized and Federated Machine Learning on Mobile Health Data","year":2021,"lang":"en","type":"review","venue":"JMIR mhealth and uhealth","topic":"Digital Mental Health Interventions","field":"Psychology","cited_by":59,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Institute of Mental Health","keywords":"Computer science; mHealth; Context (archaeology); Wearable technology; Information privacy; Wearable computer; Mobile device; Machine learning; Internet privacy; Artificial intelligence; Data science; Human–computer interaction; World Wide Web; Health care","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":[0.003656063,0.0007003206,0.0007903961,0.0005074916,0.0007772071,0.0009250609,0.001433452,0.001495661,0.002268199],"category_scores_gemma":[0.01412258,0.0003024859,0.0008692518,0.000513426,0.001625747,0.001709213,0.001653097,0.001850899,0.0002125719],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001971908,"about_ca_system_score_gemma":0.001572318,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01794914,"about_ca_topic_score_gemma":0.009778284,"domain_scores_codex":[0.9988112,0.0005747196,0.00004814914,0.0001998694,0.0001324477,0.0002337276],"domain_scores_gemma":[0.9835494,0.01369862,0.0005952942,0.0007453983,0.000891227,0.0005201278],"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":[0.0003236652,0.0002738505,0.004206546,0.00005220788,0.0000333187,0.000100666,0.000108725,0.9845033,0.0003125947,0.002794721,0.0003976904,0.006892628],"study_design_scores_gemma":[0.00005715358,0.0001731691,0.0006964335,0.00001273092,0.0000111509,0.00002105777,0.00007076444,0.9951977,0.0003497299,0.003213254,0.0001891271,0.00000772205],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"review","genre_scores_codex":[0.9083738,0.0006633993,0.08144671,0.001826301,0.000143415,0.0003906654,0.000419137,0.0003733991,0.006363092],"genre_scores_gemma":[0.9894407,0.0001052609,0.00936361,0.0001305089,0.00001048631,0.0001114381,0.000111938,0.00001099627,0.0007150695],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.01794914,"threshold_uncertainty_score":0.03568929,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2654956090264523,"score_gpt":0.5214229209617016,"score_spread":0.2559273119352493,"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."}}