文章基本信息

标题：Optimized High-Utility Itemsets Mining for Effective Association Mining Paper
其他标题：Optimized High-Utility Itemsets Mining for Effective Association Mining Paper
本地全文：下载
作者：K Rajendra Prasad
期刊名称：International Journal of Electrical and Computer Engineering
电子版ISSN：2088-8708
出版年度：2017
卷号：7
期号：5
页码：2911-2918
DOI：10.11591/ijece.v7i5.pp2911-2918
语种：English
出版社：Institute of Advanced Engineering and Science (IAES)
摘要：Association rule mining is intently used for determining the frequent itemsets of transactional database; however, it is needed to consider the utility of itemsets in market behavioral applications. Apriori or FP-growth methods generate the association rules without utility factor of items. High-utility itemset mining (HUIM) is a well-known method that effectively determines the itemsets based on high-utility value and the resulting itemsets are known as high-utility itemsets. Fastest high-utility mining method (FHM) is an enhanced version of HUIM. FHM reduces the number of join operations during itemsets generation, so it is faster than HUIM. For large datasets, both methods are very expenisve. Proposed method addressed this issue by building pruning based utility co-occurrence structure (PEUCS) for elimatination of low-profit itemsets, thus, obviously it process only optimal number of high-utility itemsets, so it is called as optimal FHM (OFHM). Experimental results show that OFHM takes less computational runtime, therefore it is more efficient when compared to other existing methods for benchmarked large datasets.
其他摘要：Association rule mining is intently used for determining the frequent itemsets of transactional database; however, it is needed to consider the utility of itemsets in market behavioral applications. Apriori or FP-growth methods generate the association rules without utility factor of items. High-utility itemset mining (HUIM) is a well-known method that effectively determines the itemsets based on high-utility value and the resulting itemsets are known as high-utility itemsets. Fastest high-utility mining method (FHM) is an enhanced version of HUIM. FHM reduces the number of join operations during itemsets generation, so it is faster than HUIM. For large datasets, both methods are very expenisve. Proposed method addressed this issue by building pruning based utility co-occurrence structure (PEUCS) for elimatination of low-profit itemsets, thus, obviously it process only optimal number of high-utility itemsets, so it is called as optimal FHM (OFHM). Experimental results show that OFHM takes less computational runtime, therefore it is more efficient when compared to other existing methods for benchmarked large datasets.
关键词：computer and Informatics; data mining;association rule mining; FHM; frequent itemsets; high-utility itemsets; HUIM