期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2015
卷号：6
期号：4
页码：3893-3896
出版社：TechScience Publications
摘要：Data Mining can be specified as an action that analyzes the data and draws out some new and nontrivial information from the large amount of databases. Traditional data mining methods have focused on finding the statistical correlations between the items that are frequently appearing in the database. High utility itemset mining is an area of research where utility based mining is a descriptive type of data mining, aimed at finding itemsets whose utilities are greater than the user specified minimum utility threshold. In this paper, we propose an algorithm, called HU-Max (High Utility Itemset Mining using Utility Matrix), for high utility itemset mining. The algorithm uses a novel strategy called Utility Matrix to store the transaction weighted utility (TWU) of set of two items. The accidental dataset is used for evaluating this algorithm.
