Apriori算法的分析与改进

The AnaLysis and Improvement of the Apriori ALgorithm

随着网上信息的不断增加,越来越多的用户迷失在信息的海洋中,如何利用有效的方法和手段从大量的信息中找出有价值、能为用户所用的知识,是数据挖掘的主要任务.由于Apriori算法在频繁模式挖掘过程中需要多次扫描数据库、算法运行时间较长,因此笔者提出一种改进的Apriori算法——FPMUDF(频繁模式挖掘利用动态函数)算法,这种算法利用事务ID进行配对,从而产生频繁项目集,减少了算法运行的时间,较好的提高了算法的性能.

With the continuous increase of online information, more and more users get lost in the sea of the information, and how to take advantage of the effective ways and means to find out valuable knowl- edge for users from the large amount of information , which is the main task of the data mining . since the Apriori algorithm in the processing of frequent pattern mining repeatedLy scan the database, having longer running time, This paper proposed an improved Apriori algorithm--FPMUDF（Frequent Pattern Mining U- sing Dynamic Function） algorithm, This algorithm used the transaction ID pairing resuLting in frequent itemsets to reduce the algorithm running time, improve the performance of the algorithm.