Abstract Views :189 |
PDF Views:2
Authors
Affiliations
1 School of Information Technology, Rajiv Gandhi Technological University, Bhopal (M.P.), IN
2 Samrat Ashok Technological Insitute (Degree), Vidisha (M.P.), IN
Source
Data Mining and Knowledge Engineering, Vol 1, No 5 (2009), Pagination: 233-238
Abstract
This paper describes a new parallel PrimeMulti algorithm for association rule mining. PrimeMulti algorithm addresses the shortcoming of previously proposed parallel buddy prima algorithm. New efficient algorithm for load balancing is also proposed in this paper. New algorithm divide transaction database equally according to the transaction length to the processors. In the Parallel PrimeMulti algorithm transaction database is represented by prime number. Less memory is requires as each transaction is replaced with the product of the equivalent prime numbers of their items. Parallel PrimeMulti algorithm works on top down as well as bottom up approach. The proposed algorithm for parallel frequent itemset mining and load balancing reduces the time and data complexity and divide transactional database efficiently for good load balancing among the processor.
Keywords
Association Rule Mining, Confidence, Frequent Item Set, Parallel Data Mining, Top-Down Approach, Support.
