{"id":"W2809070429","doi":"10.2196/medinform.7096","title":"Contextual Anonymization for Secondary Use of Big Data in Biomedical Research: Proposal for an Anonymization Matrix","year":2018,"lang":"en","type":"article","venue":"JMIR Medical Informatics","topic":"Ethics in Clinical Research","field":"Medicine","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Data anonymization; Big data; Data science; Data mining; Computer security; Information privacy","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":[0.1509771,0.001262008,0.001119947,0.008187864,0.006505024,0.01384329,0.003758669,0.00447679,0.004094273],"category_scores_gemma":[0.2249056,0.001148248,0.002901262,0.007306959,0.01808577,0.02273913,0.0134342,0.005810776,0.00137761],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.005596221,"about_ca_system_score_gemma":0.01798682,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002370184,"about_ca_topic_score_gemma":0.00255426,"domain_scores_codex":[0.83037,0.1206491,0.01666662,0.01140079,0.01839188,0.002521638],"domain_scores_gemma":[0.7545577,0.1364212,0.02374278,0.04571725,0.03507189,0.00448919],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00006593877,0.00005441757,0.002051851,0.0004734357,0.00006136149,0.0001563954,0.005461691,0.003041347,0.0007555993,0.9251183,0.007157303,0.05560248],"study_design_scores_gemma":[0.00005502791,0.0001071719,0.001253574,0.001087313,0.0001025145,0.0004055672,0.003778102,0.01668683,0.002529482,0.8894156,0.08443779,0.0001410835],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.005043146,0.0003944266,0.9611937,0.01706644,0.0004344709,0.001609428,0.0003014636,0.0003479833,0.01360891],"genre_scores_gemma":[0.09949179,0.0004276556,0.8941993,0.001345984,0.0002626754,0.00245259,0.000342114,0.0001107278,0.00136719],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.1509771,"threshold_uncertainty_score":0.798453,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.674555123165151,"score_gpt":0.6211667524373141,"score_spread":0.05338837072783686,"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."}}