Click here to view fulltext PDF

Permanent link:
https://www.ias.ac.in/article/fulltext/sadh/045/0169

Big Data; HDFS; MapReduce; Apriori; frequent itemset; association rule.

Data is increasing rapidly day by day along with the transactional database. Dividing this data and storing it in a distributed manner is an effective way for storage and retrieval. Mining such distributed data with minimum dependence between sub-problems is a crucial task. Finding frequent itemsets and corresponding association rules is a big challenge while considering the aggregation in a distributed environment. To overcome these challenges, we propose a distributed frequent itemset generation and association rule mining algorithm using MapReduce programming model. The proposed scheme generates frequent itemset and mine association rules using a synthesized distributed technique. The rules are mined in a distributed manner, and then weights are assigned to subsets of data and association rules. A proper mixture of association rules that are generated in distributed manner is done using a weighted approach. This paper presents a novel MapReduce-based synthesisapproach, which can work well over a distributed storage of large amount of data.

AMRIT PAL^{1 2} MANISH KUMAR²

1. Department of Computer Engineering and Application, GLA University, Mathura, India
2. Department of Information Technology, Indian Institute of Information Technology Allahabad, Prayagraj, India

Published

2 July 2020

Manuscript received

16 November 2017

Manuscript revised

Accepted

15 May 2020