Dynamic Request Redirection for Mining Services under Heterogeneous Client-server System

International Journal of Research and Scientific Innovation (IJRSI) | Volume V, Issue V, May 2018 | ISSN 2321–2705

Dynamic Request Redirection for Mining Services under Heterogeneous Client-server System

Prof. Ketaki Bhoyar*, Gurpreet Kaur Lohana*, Laxmi Nirmal*, Sunidhi Patil*, Rakesh Pandey*

*Department of Computer Science, DYPIEMR, Pune- 411033, Maharashtra, India

Abstract -The main purpose of developing system is to find the independent workload for that we have to find the internal task of apriori algorithm with a systematic method called Dynamical Request Redirection and Resource Provisioning .Those internal task which have independent work load has to be executed in parallel manner. Apriori algorithm is a classical association rule mining algorithm, but it has problems about frequently scanning database and generating a large number of candidate item set. To solve this problem, we have proposed Frequent Item Mining system with parallel execution techniques. A large dataset, here is divided into number of small datasets i.e. chunks. The legacy single threaded variable size chunking method leaves much to be desired in states. This is achieved by inter or intra segment assignment. The proposed system achieves the speed-up using multiple threads in heterogeneous system over existing sequential system as well as utilize the computational power which now a days provided by recently lunched multicore processor . Parallelism is used to reduce time, increase performance and fast processing. It is implementing Apriori algorithm in serial and parallel manner and performing comparison of both on the basis of varying support-count and time using multithreading java.

Keywords- Apriori, Association Rule Mining, Heterogeneous System, Candidate Item Set, Resource Provision.

I. INTRODUCTION

An Apriori is a frequent pattern mining algorithm used for finding association rules developed by Rakesh Agrawal and Ramakrishnan Srikant[4]. It operates on a list of transactions containing items (for example, products in a supermarket). Frequent instances of items with each other are mined by Frequent Item Calculation to find relationship among different items. A single transaction is called an Item set. Apriori uses a reduce support value as the main compulsion to determine whether a set of items is frequent. In start pass of the algorithm, it constructs the candidate 1-itemsets. The algorithm then generates the frequent 1-itemsets by excluding some candidate 1-itemsets if their support values are lower than the minimum support. After the algorithm finds all the frequent 1-itemsets, it joins the frequent 1-itemsets with each other to construct the candidate 2-itemsets and exclude some infrequent itemsets from the candidate 2-itemsets to create the frequent 2-itemsets. This process is replicated until no more candidate itemsets can be created from it.

Read More