Novelty Detection with Multivariate Extreme Value Statistics

Authors:
David Andrew Clifton;Samuel Hugueny;Lionel Tarassenko
Affiliations:
Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, UK OX3 7DQ;Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, UK OX3 7DQ;Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, UK OX3 7DQ
Venue:
Journal of Signal Processing Systems
Year:
2011

Citing 5
Cited 5

Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
A Mixture Approach to Novelty Detection Using Training Data with Outliers

EMCL '01 Proceedings of the 12th European Conference on Machine Learning
Pattern Recognition and Machine Learning (Information Science and Statistics)

Pattern Recognition and Machine Learning (Information Science and Statistics)
Gaussian mixture pdf in one-class classification: computing and utilizing confidence values

ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 02
Feature representation and discrimination based on Gaussian mixture model probability densities-Practices and algorithms

Pattern Recognition

Engine parameter outlier detection: verification by simulating PID controllers generated by genetic algorithm

IDA'12 Proceedings of the 11th international conference on Advances in Intelligent Data Analysis
Locating emergencies in a campus using wi-fi access point association data

Proceedings of the 2013 ACM conference on Pervasive and ubiquitous computing adjunct publication
Detecting rare events using extreme value statistics applied to epileptic convulsions in children

Artificial Intelligence in Medicine
Review: A review of novelty detection

Signal Processing
Extending the Generalised Pareto Distribution for Novelty Detection in High-Dimensional Spaces

Journal of Signal Processing Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

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.