Novelty Detection with Multivariate Extreme Value Statistics

被引：70

作者：

Clifton, David Andrew ^{[1
]}

Hugueny, Samuel ^{[1
]}

Tarassenko, Lionel ^{[1
]}

机构：

[1] Univ Oxford, Inst Biomed Engn, Dept Engn Sci, Oxford OX3 7DQ, England

来源：

JOURNAL OF SIGNAL PROCESSING SYSTEMS FOR SIGNAL IMAGE AND VIDEO TECHNOLOGY | 2011年 / 65卷 / 03期

基金：

英国工程与自然科学研究理事会; 英国惠康基金;

关键词：

Novelty detection; Extreme value theory; Patient monitoring; Multivariate statistics; Multimodal statistics; Biomedical engineering;

D O I：

10.1007/s11265-010-0513-6

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Novelty detection, or one-class classification, aims to determine if data are "normal" with respect to some model of normality constructed using examples of normal system behaviour. If that model is composed of generative probability distributions, the extent of "normality" in the data space can be described using Extreme Value Theory (EVT), a branch of statistics concerned with describing the tails of distributions. This paper demonstrates that existing approaches to the use of EVT for novelty detection are appropriate only for univariate, unimodal problems. We generalise the use of EVT for novelty detection to the analysis of data with multivariate, multimodal distributions, allowing a principled approach to the analysis of high-dimensional data to be taken. Examples are provided using vital-sign data obtained from a large clinical study of patients in a high-dependency hospital ward.

引用

页码：371 / 389

页数：19

共 28 条

[1]

[Anonymous], 2005, Extreme Value and Related Models with Applications in Engineering and Science

[2]

[Anonymous], 2006, Pattern recognition and machine learning

[3]

Bishop CM., 1995, NEURAL NETWORKS PATT

[4]

Clifton D.A., 2008, P COND MON 2008 ED U, P273

[5]

Clifton DA, 2008, AEROSP CONF PROC, P1853

[6]

Clifton DavidA., 2009, NOVELTY DETECTION EX

[7]

Coles S, 2001, An introduction to statistical modeling of extreme values, P45, DOI [DOI 10.1007/978-1-4471-3675-0, 10.1007/978-1-4471-3675-0]

[8]

Embrechts P., 2008, Modelling Extremal Events

[9] Limiting forms of the frequency distribution of the largest or smallest member of a sample [J].

Fisher, RA ;

Tippett, LHC .

PROCEEDINGS OF THE CAMBRIDGE PHILOSOPHICAL SOCIETY, 1928, 24 :180-190

[10]

Hann A., 2008, THESIS U OXFORD

← 1 2 3 →