IRMA-International.org: Creator of Knowledge
Information Resources Management Association
Advancing the Concepts & Practices of Information Resources Management in Modern Organizations

A Distributed Algorithm for Mining Fuzzy Association Rules in Traditional Databases

A Distributed Algorithm for Mining Fuzzy Association Rules in Traditional Databases
View Sample PDF
Author(s): Wai-Ho Au (Microsoft Corporation, USA)
Copyright: 2009
Pages: 21
Source title: Database Technologies: Concepts, Methodologies, Tools, and Applications
Source Author(s)/Editor(s): John Erickson (University of Nebraska, Omaha, USA)
DOI: 10.4018/978-1-60566-058-5.ch148

Purchase

View A Distributed Algorithm for Mining Fuzzy Association Rules in Traditional Databases on the publisher's website for pricing and purchasing information.

Abstract

The mining of fuzzy association rules has been proposed in the literature recently. Many of the ensuing algorithms are developed to make use of only a single processor or machine. They can be further enhanced by taking advantage of the scalability of parallel or distributed computer systems. The increasing ability to collect data and the resulting huge data volume make the exploitation of parallel or distributed systems become more and more important to the success of fuzzy association rule mining algorithms. This chapter proposes a new distributed algorithm, called DFARM, for mining fuzzy association rules from very large databases. Unlike many existing algorithms that adopt the support-confidence framework such that an association is considered interesting if it satisfies some user-specified minimum percentage thresholds, DFARM embraces an objective measure to distinguish interesting associations from uninteresting ones. This measure is defined as a function of the difference in the actual and the expected number of tuples characterized by different linguistic variables (attributes) and linguistic terms (attribute values). Given a database, DFARM first divides it into several horizontal partitions and assigns them to different sites in a distributed system. It then has each site scan its own database partition to obtain the number of tuples characterized by different linguistic variables and linguistic terms (i.e., the local counts), and exchange the local counts with all the other sites to find the global counts. Based on the global counts, the values of the interestingness measure are computed, and the sites can uncover interesting associations. By repeating this process of counting, exchanging counts, and calculating the interestingness measure, it unveils the underlying interesting associations hidden in the data. We implemented DFARM in a distributed system and used a popular benchmark data set to evaluate its performance. The results show that it has very good size-up, speedup, and scale-up performance. We also evaluated the effectiveness of the proposed interestingness measure on two synthetic data sets. The experimental results show that it is very effective in differentiating between interesting and uninteresting associations.

Related Content

Renjith V. Ravi, Mangesh M. Ghonge, P. Febina Beevi, Rafael Kunst. © 2022. 24 pages.
Manimaran A., Chandramohan Dhasarathan, Arulkumar N., Naveen Kumar N.. © 2022. 20 pages.
Ram Singh, Rohit Bansal, Sachin Chauhan. © 2022. 19 pages.
Subhodeep Mukherjee, Manish Mohan Baral, Venkataiah Chittipaka. © 2022. 17 pages.
Vladimir Nikolaevich Kustov, Ekaterina Sergeevna Selanteva. © 2022. 23 pages.
Krati Reja, Gaurav Choudhary, Shishir Kumar Shandilya, Durgesh M. Sharma, Ashish K. Sharma. © 2022. 18 pages.
Nwosu Anthony Ugochukwu, S. B. Goyal. © 2022. 23 pages.
Body Bottom