摘要
基于任务分解的策略,提出多支撑向量网络算法.由于每个子网络只完成相应的子任务,因而多支撑向量网络可以解决更为复杂的学习任务.同时整个系统更容易理解和修正并且具有更好的系统性能.仿真试验结果表明该方法是有效的.
Based on the divided-and-conquer strategy, multiple support vector machine is proposed. Multiple support vector machine can perform more complex tasks because every support vector machine only finish the subtask. At the same time, It can make an overall system easier to understand and modify. Computation results show that the method in this paper is availability.
出处
《小型微型计算机系统》
CSCD
北大核心
2006年第5期846-848,共3页
Journal of Chinese Computer Systems
关键词
聚类方法
多支撑向量网络
二次规划
clustering algorithm
multiple support vector machine
quadratic programming