多分类支撑向量机综述

支撑向量机的研究是当前人工智能领域的研究热点。传统的支撑向量机是两类的分类器,如何将其有效地推广到多类问题仍是一个有待研究的问题。在本文中,作者对目前已有的多分类支撑向量机算法,从基本思想、实现的方法、特点以及存在的问题上做了详细的介绍,以期对今后的研究有所启发。

Multi-Class Support Vector Machine Summarazation

At present,studies on support vector machines are hot topics in the field of artificial intelligence.Traditional support vector machines were classifiers for binary classification.How to extend it for multi-class classification effectively is still an on-going research issue.In this paper,we will detailedly introduce several existing multi-class support vector machine algorithms on basic thinkings,implements,characteristics and existent ploblems. The author hopes that this paper can give investigators some illumination in their present-day investigation.