{"id":"W2000219982","doi":"10.1109/icde.2014.6816641","title":"Scalable distance-based outlier detection over high-volume data streams","year":2014,"lang":"en","type":"article","venue":"","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":119,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Ministère de l'Économie, de la Science et de l'Innovation - Québec; National Science Foundation","keywords":"Anomaly detection; Outlier; Computer science; Scalability; Data mining; Data stream mining; Ranging; Data point; Cluster analysis; Scale (ratio); Big data; Process (computing); Credit card fraud; Object (grammar); Artificial intelligence; Credit card; Database","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001782854,0.001058144,0.001639043,0.001849033,0.0006515619,0.001578641,0.002570368,0.0009874156,0.0005201267],"category_scores_gemma":[0.01176577,0.0004263011,0.00067442,0.002399881,0.0007211586,0.002959091,0.002184487,0.001642994,0.0004195869],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007274096,"about_ca_system_score_gemma":0.001023352,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002567759,"about_ca_topic_score_gemma":0.002331716,"domain_scores_codex":[0.9977615,0.0003179946,0.0001756414,0.0004832643,0.001118221,0.0001433508],"domain_scores_gemma":[0.9931646,0.003117854,0.00108947,0.00104794,0.001279365,0.0003007245],"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":[0.0005174266,0.0003397706,0.01322256,0.0002677665,0.0001996851,0.0004810131,0.0003379146,0.5179365,0.02880296,0.01090559,0.003642751,0.4233461],"study_design_scores_gemma":[0.000008461243,0.00004974016,0.0007485806,0.000005292043,0.000007536193,0.00009432131,0.00005072662,0.9899243,0.003607171,0.004923393,0.0005694919,0.00001094661],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02650595,0.0002501617,0.971242,0.0001469337,0.00004382525,0.00005374485,0.0001018097,0.001316147,0.0003396134],"genre_scores_gemma":[0.5868176,0.000366104,0.4104693,0.0001053502,0.0001503094,0.0001155552,0.0006945028,0.0001565823,0.001124655],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002570368,"threshold_uncertainty_score":0.00942874,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01346393922891818,"score_gpt":0.2376409428860461,"score_spread":0.2241770036571279,"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."}}