{"id":"W2292231671","doi":"10.1007/s10115-015-0906-8","title":"Managing dimensionality in data privacy anonymization","year":2015,"lang":"en","type":"article","venue":"Knowledge and Information Systems","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Calgary","funders":"","keywords":"Curse of dimensionality; Computer science; Data mining; Closeness; Process (computing); Cluster analysis; Variety (cybernetics); Machine learning; Mathematics; Artificial intelligence","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":[],"consensus_categories":[],"category_scores_codex":[0.01564143,0.000640392,0.002217445,0.002186804,0.004085601,0.009264796,0.002298712,0.002609704,0.001630698],"category_scores_gemma":[0.06057881,0.0009735006,0.001506044,0.004840464,0.007638927,0.02364261,0.01131511,0.005955392,0.0004783524],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002830898,"about_ca_system_score_gemma":0.004192434,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001435431,"about_ca_topic_score_gemma":0.001185717,"domain_scores_codex":[0.9721124,0.01505653,0.001847505,0.002780304,0.006862339,0.001340824],"domain_scores_gemma":[0.9323665,0.0266008,0.003880511,0.03289529,0.003467994,0.0007889242],"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":[0.0002902078,0.0001481972,0.002821769,0.0002662113,0.0001410163,0.0002136536,0.001487285,0.05890441,0.001943897,0.8255437,0.005608246,0.1026314],"study_design_scores_gemma":[0.00002289621,0.00003771003,0.000400183,0.00007066378,0.00006223562,0.0002737582,0.0006079162,0.1403836,0.00473957,0.842837,0.01052421,0.00004020706],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02792364,0.001589029,0.956005,0.007636402,0.000223205,0.0001374794,0.0003700867,0.0003257009,0.005789569],"genre_scores_gemma":[0.7464541,0.002123264,0.2458334,0.0009601437,0.0004976075,0.0002221665,0.0004610342,0.0001123453,0.003336011],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01564143,"threshold_uncertainty_score":0.08272082,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07068525535148379,"score_gpt":0.3013851363574971,"score_spread":0.2306998810060134,"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."}}