期刊名称：International Journal of Computer Science & Technology
印刷版ISSN：2229-4333
电子版ISSN：0976-8491
出版年度：2015
卷号：6
期号：3
页码：175-178
语种：English
出版社：Ayushmaan Technologies
摘要：Traditional pattern mining algorithms may not discover a number of most beneficial, high priced patterns, due to their lower support. These algorithms reflect only statistical correlation, but it does not reflect semantic significance of the pattern. Although a number of relevant algorithms have been proposed in recent years, they incur the problem of producing a large number of candidate item-sets for High Utility item-sets. Such a large number of candidate item-sets degrade the mining performance in terms of execution time andspace requirement. Also the previous algorithms do not consider the impact of item sets with high tolls. The Proposed strategies in this work can not only decrease the overestimated utilities of potential high utility item sets but greatly reduce the number of candidates. Different types of both real and synthetic data sets are used in a series of experiments to the performance of the proposed algorithm with state-of-the-art utility mining algorithms.Experimental results show that these algorithms outperform other algorithms substantially in term of execution time, especially when databases contain lots of long transactions or low minimum utility thresholds are set.
关键词：Data mining;Association rule mining;Support and Confidence,
