{"id":"W3196335171","doi":"10.20517/jsss.2020.20","title":"Information theoretic-based privacy risk evaluation for data anonymization","year":2022,"lang":"en","type":"article","venue":"Journal of Surveillance Security and Safety","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Metric (unit); Computer science; Inference; Data anonymization; Data publishing; Identification (biology); Data mining; Process (computing); Information sensitivity; Information privacy; Publishing; Artificial intelligence; Computer security","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":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.00960126,0.00009455599,0.0001824381,0.0001665305,0.0005216785,0.0001645081,0.008691585,0.00005398036,0.00001966478],"category_scores_gemma":[0.02248145,0.00008754744,0.00004062285,0.000350827,0.00006472789,0.002076556,0.01345818,0.000337493,5.262538e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001451761,"about_ca_system_score_gemma":0.0002881203,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000155099,"about_ca_topic_score_gemma":0.00001653383,"domain_scores_codex":[0.9980248,0.0004342216,0.0005206664,0.000161907,0.0007034456,0.0001549516],"domain_scores_gemma":[0.9955288,0.0004883895,0.0008903951,0.002615405,0.0004367269,0.00004030098],"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":[0.001745899,0.0004585603,0.05578414,0.0003460374,0.0002253093,0.000007169658,0.003848684,0.01438749,0.00006315819,0.0311691,0.1397552,0.7522092],"study_design_scores_gemma":[0.001155247,0.000166054,0.005593552,0.00000871954,0.00001039207,0.0000200619,0.00006866067,0.794347,0.00003886014,0.1815677,0.01692406,0.00009965846],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02637421,0.0004235768,0.9552367,0.01625962,0.00055786,0.0004398137,0.0005220225,0.000073617,0.0001125992],"genre_scores_gemma":[0.9228165,0.0003024981,0.07643642,0.0001532216,0.00003946199,0.000009066105,0.0002383542,0.000004120735,3.589566e-7],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8964423,"threshold_uncertainty_score":0.9966719,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02722922051687764,"score_gpt":0.2842230224223194,"score_spread":0.2569938019054417,"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."}}