A new method for anomaly detection based on non-convex boundaries with random two-dimensional projections
View/ Open
Use this link to cite
http://hdl.handle.net/2183/31882
Except where otherwise noted, this item's license is described as This accepted manuscript version is made available under the CC Attribution-NonCommercialNoDerivatives 4.0 International license: http://creativecommons.org/licenses/by-nc-nd/4.0
Collections
- GI-CTC - Artigos [80]
Metadata
Show full item recordTitle
A new method for anomaly detection based on non-convex boundaries with random two-dimensional projectionsAuthor(s)
Date
2020-08-19Citation
E. Jove, J.-L. Casteleiro-Roca, H. Quintián, J.-A. Méndez-Pérez, J.L. Calvo-Rolle, A new method for anomaly detection based on non-convex boundaries with random two-dimensional projections, Information Fusion. 65 (2021) 50–57. https://doi.org/10.1016/j.inffus.2020.08.011
Abstract
[Abstract] The implementation of anomaly detection systems represents a key problem that has been focusing the efforts of scientific community. In this context, the use one-class techniques to model a training set of non-anomalous objects can play a significant role. One common approach to face the one-class problem is based on determining the geometric boundaries of the target set. More specifically, the use of convex hull combined with random projections offers good results but presents low performance when it is applied to non-convex sets. Then, this work proposes a new method that face this issue by implementing non-convex boundaries over each projection. The proposal was assessed and compared with the most common one-class techniques, over different sets, obtaining successful results.
Keywords
One-class
Anomaly detection
Projection methods
Convex hull
Boundary
Limits
Anomaly detection
Projection methods
Convex hull
Boundary
Limits
Editor version
Rights
This accepted manuscript version is made available under the CC Attribution-NonCommercialNoDerivatives 4.0 International license: http://creativecommons.org/licenses/by-nc-nd/4.0
ISSN
1566-2535