摘要
传统算法无法避免云计算环境下冗余数据特征的动态变化性和特征多样性,从而降低了冗余数据分类的效果。为此,提出一种基于优化相关向量机的冗余数据分类方法。利用相关向量机建立冗余数据分类模型,并获得需要确定的参数,将参数看作粒子,构建初始粒子群,通过粒子群算法进行迭代寻优,获得最优分类模型的参数,从而实现云计算环境下冗余数据的准确分类。
The traditional algorithms can't eliminate dynamic change and diversity of redundant data features in cloud computing environment, which reduces the redundant data classification effect. In view of this problem, this paper puts forward a kind of redundant data classification method based on optimal relevance vector machine. In this method, the relevance vector machine is used to establish redundant data classification model and obtain the parameters to be determined, and the determined parameters are taken as particles to build the inifal particle swarm. The particle swarm optimization algorithm is used to implement iterative optimization and obtain the optimal classification model parameters, so as to realize accurate classification of redundant data in cloud computing environment.
出处
《计算机与网络》
2015年第20期68-71,共4页
Computer & Network
关键词
优化相关向量机
云计算环境
冗余数据
分类
optimal relevance vector machine
cloud computing environment
redundant data
classification