{"id":"W4283008962","doi":"10.1371/journal.pone.0269097","title":"Measuring re-identification risk using a synthetic estimator to enable data sharing","year":2022,"lang":"en","type":"article","venue":"PLoS ONE","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"ca_institutions":"Children's Hospital of Eastern Ontario; University of Ottawa; University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs; FedDev Ontario; University of Toronto; Compute Canada","keywords":"Estimator; Computer science; Population; Statistics; Copula (linguistics); Data mining; Econometrics; Mathematics; Medicine","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.001697958,0.0001261073,0.0001817853,0.0002451567,0.0006617299,0.0003474454,0.04873611,0.00003451241,0.00004706427],"category_scores_gemma":[0.03073828,0.0001532866,0.0000189868,0.0008554245,0.00002670294,0.001006181,0.2570976,0.0003525463,0.00007316372],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002783998,"about_ca_system_score_gemma":0.00005974751,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002011839,"about_ca_topic_score_gemma":0.000008909697,"domain_scores_codex":[0.99737,0.00009155512,0.0003040078,0.001089907,0.0007763568,0.0003681398],"domain_scores_gemma":[0.9796404,0.0001359904,0.0001826133,0.01990861,0.00005817967,0.00007422618],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007807201,0.005671116,0.03112927,0.0008558424,0.001158853,0.0002337772,0.001766989,0.01641361,0.8329564,0.004949327,0.06626302,0.03852371],"study_design_scores_gemma":[0.0000659583,0.00002080234,0.0001730926,0.0001178402,0.00004300398,0.000006324052,0.00004296462,0.9368727,0.04140987,0.02092777,0.0001245485,0.0001951151],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.4388324,0.0002573048,0.5493962,0.008217234,0.0002759322,0.0006488232,0.0002342665,0.001881387,0.000256411],"genre_scores_gemma":[0.464422,0.000009157683,0.5353891,0.00004068535,0.00002187611,0.00005610264,0.00001688801,0.00001749432,0.00002667907],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9204591,"threshold_uncertainty_score":0.9774262,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2833191170127614,"score_gpt":0.2998187574183161,"score_spread":0.01649964040555468,"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."}}