云计算环境下基于MapReduce并行的Apriori算法优化研究

根据MapReduce模型并行运行实现的特点,针对可扩展性差的传统Apriori的特点和传统Apriori算法,采用了＂云＂强大的廉价计算处理方式和关联规则挖掘算法,改进提高Apriori算法的运算效率。通过改进在云计算环境下MapReduce编程框架,并且结合验证MR-Apriori算法的实验为基础,这对传统意义上的Apriori算法在数据挖掘过程中所出现的客观问题进行处理,从而真正意义上的完成了本文研究的基于MapReduce并行的Apriori算法的扩展性提升的目标,并且表明了元计算技术结合关联规则挖掘算法的可能性。

Optimization of Apriori algorithm based on parallel MapReduce in cloud computing environment

Based on the characteristics of MapReduce model running in parallel, consider poor scalability characteristics of conventional and traditional Apriori algorithm, this paper adoptes a ＂cloud＂ of cheap computing power handling and association rule mining algorithm, Improves Apriori algorithm to improve operational efficiency. By improving the environment in the cloud MapReduce programming framework, and combined with MR-Apriori algorithm validation experiments, MR-Apriori algorithm for parallel improvements, the traditional Apriori mining to resolve the problems encountered, implemented based on MapReduce parallel Apriori algorithm is highly scalable, and shows the element technology combined with the likelihood of association rule mining algorithm.