An alternative to SVM Method for Data Classification

TL;DR

This paper proposes a new classification method that addresses SVM's limitations in high-dimensional, multi-class, unbalanced, and dynamic data scenarios by using minimum distance to optimal subspaces.

Contribution

The paper introduces an alternative to SVM based on minimum distance to optimal subspaces, improving processing time and robustness in complex classification tasks.

Findings

Comparable performance to SVM

Improved processing time and robustness

Better handling of high-dimensional and unbalanced data

Abstract

Support vector machine (SVM), is a popular kernel method for data classification that demonstrated its efficiency for a large range of practical applications. The method suffers, however, from some weaknesses including; time processing, risk of failure of the optimization process for high dimension cases, generalization to multi-classes, unbalanced classes, and dynamic classification. In this paper an alternative method is proposed having a similar performance, with a sensitive improvement of the aforementioned shortcomings. The new method is based on a minimum distance to optimal subspaces containing the mapped original classes.