期刊名称：International Journal of Computer Technology and Applications
电子版ISSN：2229-6093
出版年度：2011
卷号：2
期号：5
页码：1256-1262
出版社：Technopark Publications
摘要：In recent days Data mining techniques have been widely used in various applications. One of the most important applications in data mining is association rule mining. For hardware implementation of Apriori-based association rule mining we have to load candidate item sets and a database into the hardware. As the hardware architecture capacity is fixed, when the number of items or the number of candidate item sets in database is larger than the hardware capacity, the items are loaded into the hardware separately. Which increases the time complexity to those steps that require to load candidate item sets or database items into the hardware which is proportional to the number of candidate item sets multiplied by the number of items in the database. As the time complexity is increasing because of many candidate item sets and use of large database , which is finally reflecting the performance bottleneck. In this paper, we propose a HAsh-based and PiPelIned (abbreviated as HAPPI) architecture to enhance the implementation of association rule mining on hardware. Hence, we can effectively decrease the frequency of loading the database into the hardware. HAPPI solves the bottleneck problem in a priori-based hardware schemes. Along with this hashing we are including here the privacy preservation for the sensitive data that is processing in the data mining. It is a common problem that is being faced by all Data Mining Techniques.
关键词：Apriori-based; Privacy; HAPPI; Items; Trimming; Information
