Graph-Embedded Subspace Support Vector Data Description

Abstract

In this paper, we propose a novel subspace learning framework for one-class classification. The proposed framework presents the problem in the form of graph embedding. It includes the previously proposed subspace one-class techniques as its special cases and provides further insight on what these techniques actually optimize. The framework allows to incorporate other meaningful optimization goals via the graph preserving criterion and reveals a spectral solution and a spectral regression-based solution as alternatives to the previously used gradient-based technique. We combine the subspace learning framework iteratively with Support Vector Data Description applied in the subspace to formulate Graph-Embedded Subspace Support Vector Data Description. We experimentally analyzed the performance of newly proposed different variants. We demonstrate improved performance against the baselines and the recently proposed subspace learning methods for one-class classification.

25 pages manuscript (3 tables, 4 figures), 63 pages supplementary material (43 tables, 34 figures). The manuscript and supplementary material are combined as a single .pdf (88 pages) file

References in corpus (5)

• Multimodal Subspace Support Vector Data Description
• Ellipsoidal Subspace Support Vector Data Description
• Average Localised Proximity: A new data descriptor with good default one-class classification performance
• Boosting rare benthic macroinvertebrates taxa identification with one-class classification
• Saliency-based Weighted Multi-label Linear Discriminant Analysis

Cited by in corpus (3)

• Audio-based Anomaly Detection in Industrial Machines Using Deep One-Class Support Vector Data Description
• Hyperspectral Image Analysis with Subspace Learning-based One-Class Classification
• Improved Active Fire Detection using Operational U-Nets