Call For Paper Volume:4 Issue:10 Oct'2017 |

Frequent Pattern Mining Based on Maximum and Minimum Count Approach for Different Computing Environments

Publication Date : 21/03/2015



Author(s) :

Parag Moteria , Dr Y.R.Ghodasara.


Volume/Issue :
Volume 2
,
Issue 3
(03 - 2015)



Abstract :

Parag Moteria Ph.D. Scholar, School of Computer Science, R K University, Rajkot and Assistant Professor, MCA Department, ISTAR, Vallabh Vidyanagar Data mining is the process of finding interesting trends or patterns in transactional datasets to discover knowledge and helps in decision making process. Frequent patterns are frequent data set in transactional dataset. Discover interesting associated frequent items among large itemsets by specifying a user defined support threshold without any domain knowledge leads small or large numbers of uninterested results, is not appropriate. Without specifying minimum user defined support threshold is high cost effective process, but it helps to conclude interesting frequent itemsets without any domain knowledge. We propose new algorithm, RCRS (Reduce Candidate Itemsets and Reduce Scanning of dataset) based on Itemwise Cardinality Matrix, Maximum Item Count, and Minimum Item Count approach implement on three different types of computing environment. This RCRS algorithm has multiple features: only single scan is required to form Itemwise Cardinality Matrix, Maximum Item Count and Minimum Item Count derived from Itemwise Cardinality Matrix, Minimum Item Count table helps to determine implicit minimum support count threshold, and Maximum Item Count helps to reduce items in candidate set and numbers of scan to determine frequent itemsets in transactional dataset both. Centralized transactional dataset computing on demand by distributed node and achieve scalability with faster result using distributed computing environment. Experiment shows RCRS is more efficient than Apriori.


No. of Downloads :

31


Indexing

Web Design MymensinghPremium WordPress ThemesWeb Development

Frequent Pattern Mining Based on Maximum and Minimum Count Approach for Different Computing Environments

March 20, 2015