{"id":"W4410692212","doi":"10.1007/s10115-025-02417-2","title":"Providing and evaluating a model for big data anonymization streams by using in-memory processing","year":2025,"lang":"en","type":"article","venue":"Knowledge and Information Systems","topic":"Privacy-Preserving Technologies in Data","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"Regional Municipality of Niagara","funders":"","keywords":"Computer science; Big data; Data stream mining; Data mining; STREAMS; Data anonymization; Data science; Artificial intelligence; Information privacy; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004200311,0.001032779,0.0008916215,0.001086114,0.001215505,0.003406567,0.002492108,0.001459117,0.001737957],"category_scores_gemma":[0.01115795,0.0004105819,0.0009348727,0.001355937,0.001203185,0.004907236,0.001533148,0.001861955,0.0006400116],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002800046,"about_ca_system_score_gemma":0.002715582,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008318335,"about_ca_topic_score_gemma":0.007460958,"domain_scores_codex":[0.9975433,0.0008302453,0.000144368,0.000539962,0.0006739047,0.000268261],"domain_scores_gemma":[0.994991,0.002391151,0.0004159378,0.0009950255,0.0009996226,0.0002072179],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003459089,0.000194448,0.002172078,0.00009382809,0.00004921802,0.0001117003,0.0001882691,0.9168168,0.002286604,0.03575407,0.0033523,0.03863468],"study_design_scores_gemma":[0.000005367132,0.0000146862,0.00005252775,0.000002809178,0.000003614197,0.00001651601,0.00002519681,0.9929912,0.0008140546,0.005573289,0.0004963279,0.000004417443],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03473398,0.0001925921,0.9599126,0.0007986446,0.0001248945,0.0002166751,0.0002829995,0.001173964,0.00256368],"genre_scores_gemma":[0.5500241,0.0005094272,0.4444589,0.0002672859,0.0001405038,0.0003997266,0.001036232,0.0002188607,0.002944931],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008318335,"threshold_uncertainty_score":0.02221364,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1086372386100229,"score_gpt":0.3474931435090034,"score_spread":0.2388559048989805,"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."}}