基于Web Service技术分布式并行数据挖掘的研究与实现

Reserch and Realization of Distributed and Parallel Data Mining Based on Web Service

主要介绍基于Web Service技术的一个数据挖掘系统,在一个关联规则挖掘的并行算法—CD算法的基础上,结合一种基于动态数据集划分的并行关联规则挖掘算法,利用动态方式分配数据量,使每个处理器获得相同多的数据集,解决在网络中大量分散的数据因通信等问题而引起的负载平衡,从而提高了数据挖掘效率。

The paper gives a distributed data mining system based on Web Services. On the basis of CD algorithm,the paper presents a parallel algorithm for mining association rules based on dynamic dataset partition. By uning dynamic method to allocate data,a processor can obtain same dataset. Since it solves the load balance better because of the problems of distributed -data and communication,improves the efficiency of data mining.