{"id":"W2078902626","doi":"10.2196/jmir.2001","title":"De-identification Methods for Open Health Data: The Case of the Heritage Health Prize Claims Dataset","year":2012,"lang":"en","type":"article","venue":"Journal of Medical Internet Research","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":71,"is_retracted":false,"has_abstract":true,"ca_institutions":"Privacy Analytics (Canada); Agricultural Research Institute of Ontario; University of Ottawa","funders":"","keywords":"Health Insurance Portability and Accountability Act; Identification (biology); Context (archaeology); Computer science; Competition (biology); Public health; Big data; Data science; Data mining; Actuarial science; Business; Computer security; Confidentiality; Medicine; Geography","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.04225891,0.000711493,0.0007644661,0.002312717,0.002262522,0.002588537,0.00283634,0.003416651,0.0009756557],"category_scores_gemma":[0.0865389,0.0003006467,0.001401548,0.002882643,0.002555046,0.004600654,0.003723928,0.004246905,0.000369059],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002659037,"about_ca_system_score_gemma":0.002249097,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006412605,"about_ca_topic_score_gemma":0.007054026,"domain_scores_codex":[0.974111,0.01194577,0.001901274,0.002919814,0.0082071,0.0009150998],"domain_scores_gemma":[0.9024991,0.06320602,0.007058742,0.02030342,0.005586972,0.001345838],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002886665,0.001615875,0.1422467,0.001361769,0.000889476,0.003560547,0.002808536,0.3414279,0.006676102,0.1463531,0.07091244,0.2792608],"study_design_scores_gemma":[0.0002798365,0.0003455239,0.02243953,0.0003090478,0.00008553589,0.002389492,0.00125309,0.8192822,0.01018921,0.1085278,0.03477717,0.0001214257],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5528274,0.003185848,0.3875903,0.02410863,0.0008823601,0.001796233,0.01612697,0.001928163,0.01155411],"genre_scores_gemma":[0.6640168,0.0004422623,0.3224137,0.00116142,0.0002885581,0.0004938156,0.009587686,0.0001226462,0.001473103],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9577411,"threshold_uncertainty_score":0.2234891,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3991933750815713,"score_gpt":0.5972340287401202,"score_spread":0.1980406536585489,"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."}}