Graph-Embedded Subspace Support Vector Data Description
This work addresses one-class classification problems, which are important for anomaly detection and outlier identification in various domains, but it appears to be an incremental improvement over existing subspace techniques.
The authors tackled the problem of one-class classification by proposing a novel subspace learning framework that incorporates graph embedding, which generalizes existing subspace one-class techniques and reveals their underlying optimization principles. They demonstrated improved performance against baseline methods and recently proposed subspace learning approaches for one-class classification.
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.