Abstract:
Frequent Item set mining is widely used in market basket analysis, web link analysis, click stream analysis and drug design. The major concern of these industries is faster processing of a very large amount of data. Frequent Item set mining can be performed using Apriori, FP-tree, and Eclat algorithms. Among these, Eclat provides better efficiency when it is about processing a large amount of data. To process very large amount of data, there is a variation of Eclat called Map-Reduce Eclat. It maps large amount of transactional data and then reduces to generate the input for the Eclat algorithm. This saves the time for Eclat algorithm to process the transactional data. But Map-Reduce algorithm cannot handle memory size and improve only scalability. Than Map-reduce working on only limited dataset. It is applicable only to reduce the map data. It reduces access time that is why this dissertation proposes parallel Eclat algorithm to improve work efficiency, scalability and handle the large amount of dataset. This dissertation work discusses the comparison of existing and proposed algorithm.