Academic Journals Database
Disseminating quality controlled scientific knowledge

Optimized Frequent Pattern Mining for Classified Data Sets

ADD TO MY LIST
 
Author(s): A Raghunathan | K Murugesan

Journal: International Journal of Computer Applications
ISSN 0975-8887

Volume: 1;
Issue: 27;
Start page: 20;
Date: 2010;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Data mining | association rule | Apriori algorithm | transactions | frequent items | itemsets.

ABSTRACT
Mining frequent patterns in data is a useful requirement in several applications to guide futuredecisions. Association rule mining discovers interesting relationships among a large set ofdata items. Several association rule mining techniques exist, with the Apriori algorithm beingcommon. Numerous algorithms have been proposed for efficient and fast association rulemining in data bases, but these seem to only look at the data as a set of transactions, eachtransaction being a collection of items. The performance of the association rule techniquemainly depends on the generation of candidate sets. In this paper we present a modifiedApriori algorithm for discovering frequent items in data sets that are classified intocategories, assuming that a transaction involves maximum one item being picked up fromeach category. Our specialized algorithm takes less time for processing on classified data setsby optimizing candidate generation. More importantly, the proposed method can be used fora more efficient mining of relational data bases.
Save time & money - Smart Internet Solutions      Why do you need a reservation system?