摘要
基于二进制编码技术,提出一种快速多分类支持向量机实现策略.首先描述编码技术的实现思想,进一步给出避免多分类器中各个支持向量机的正负类样本数目不均衡的方法.并提出寻找多类别之间的最佳划分的策略,使该分类器在牺牲较小精度的情况下,具有较快的分类速度,适合应用于实时或在线分类系统中.最后根据实验结果对这种多分类器系统的性能进行评述.
A binary encoding based fast multi-class support vector machine(SVM) is introduced. How to avoid the uneven class size of each SVM in the multi-classification system is discussed based on the encoding method. Then the strategy of searching the optimal division of different classes is proposed. Thus, with little loss of accuracy the system has a higher classification speed than the traditional ones. Therefore, the classifier is suitable for real time or online systems. Finally, the introduced classification system is evaluated by experiments.
出处
《模式识别与人工智能》
EI
CSCD
北大核心
2007年第3期301-307,共7页
Pattern Recognition and Artificial Intelligence
关键词
支持向量机
多分类
二进制编码
Support Vector Machine, Multi-Class, Binary Encoding