White Rose University Consortium logo
University of Leeds logo University of Sheffield logo York University logo

Information theoretic novelty detection

Filippone, M. and Sanguinetti, G. (2010) Information theoretic novelty detection. Pattern Recognition, 43 (3). pp. 805-814. ISSN 0031-3203

Full text available as:
[img] Text
Filippone3.pdf

Download (569Kb)

Abstract

We present a novel approach to online change detection problems when the training sample size is small. The proposed approach is based on estimating the expected information content of a new data point and allows an accurate control of the false positive rate even for small data sets. In the case of the Gaussian distribution, our approach is analytically tractable and closely related to classical statistical tests. We then propose an approximation scheme to extend our approach to the case of the mixture of Gaussians. We evaluate extensively our approach on synthetic data and on three real benchmark data sets. The experimental validation shows that our method maintains a good overall accuracy, but significantly improves the control over the false positive rate.

Item Type: Article
Copyright, Publisher and Additional Information: © 2009 Elsevier. This is an author produced version of a paper publication in Pattern Recognition. Uploaded in accordance with the publisher's self-archiving policy
Keywords: Novelty Detection; Information Theory; Mixture of Gaussians; Density Estimation
Institution: The University of Sheffield
Academic Units: The University of Sheffield > Faculty of Engineering (Sheffield) > Department of Computer Science (Sheffield)
Depositing User: Miss Anthea Tucker
Date Deposited: 30 Oct 2009 10:14
Last Modified: 08 Feb 2013 16:59
Published Version: http://dx.doi.org/10.1016/j.patcog.2009.07.002
Status: Published
Publisher: Elsevier
Refereed: Yes
Identification Number: 10.1016/j.patcog.2009.07.002
URI: http://eprints.whiterose.ac.uk/id/eprint/10046

Actions (repository staff only: login required)