{"id":"W2484962737","doi":"10.1016/j.jbi.2016.07.015","title":"A unified framework for evaluating the risk of re-identification of text de-identification tools","year":2016,"lang":"en","type":"article","venue":"Journal of Biomedical Informatics","topic":"Electronic Health Records Systems","field":"Health Professions","cited_by":28,"is_retracted":false,"has_abstract":true,"ca_institutions":"Children's Hospital of Eastern Ontario; Privacy Analytics (Canada); University of Ottawa","funders":"Natural Sciences and Engineering Research Council of Canada; Mitacs; University of Ottawa","keywords":"Identification (biology); Identifier; Computer science; Context (archaeology); Data mining; Set (abstract data type); Information retrieval","routes":{"ca_aff":true,"ca_fund":true,"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"],"consensus_categories":[],"category_scores_codex":[0.02117333,0.0001076342,0.0004444896,0.0002319571,0.0002841583,0.00001255651,0.0004845624,0.0003384432,0.00007308336],"category_scores_gemma":[0.02300234,0.00005569825,0.0001659519,0.0003942489,0.0001876082,0.0003277118,0.00004573483,0.0006498552,0.00002172222],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003744882,"about_ca_system_score_gemma":0.00189138,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002539832,"about_ca_topic_score_gemma":0.000006601257,"domain_scores_codex":[0.992753,0.0008675542,0.005045177,0.00006245543,0.0008651243,0.0004066685],"domain_scores_gemma":[0.9801405,0.00895324,0.008968635,0.0004151521,0.001350434,0.000172037],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001540171,0.0005937474,0.01382736,0.01183627,0.00087925,8.861841e-7,0.07949229,0.00009365365,0.07562,0.03322828,0.03235181,0.7505363],"study_design_scores_gemma":[0.02463722,0.01500342,0.1322238,0.04207252,0.002802038,0.0001686822,0.1571974,0.1445592,0.03749046,0.3156612,0.1266069,0.001577125],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3957617,0.000119422,0.5985281,0.003420409,0.001040531,0.0009643604,0.00007432262,0.000009718288,0.00008145515],"genre_scores_gemma":[0.9801067,0.0004514224,0.01846863,0.0002145523,0.0004456067,0.00006055582,0.00000666897,0.00001666951,0.000229193],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7489592,"threshold_uncertainty_score":0.9852273,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1309630934617878,"score_gpt":0.4942414625981897,"score_spread":0.3632783691364019,"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."}}