Parallel implementation of Apriori algorithm on Multicore System for Retail Market

Data Mining is a process of analyzing data and revealing the interesting patterns which are hidden. Association Rule Mining is a key technique of data mining. This technique works on discovering interesting relationships. Association rules are generated using Apriori Algorithm. The set of data includes a number of items which are called transactions. The work of this algorithm is to produce frequent itemsets from the transactional databases based upon the minimum support value. The output of Apriori Algorithm is sets of association rules that provide us the frequency of items that are contained in sets of data which provide us the hidden pattern and general trends. This will be helpful for the retailer be familiar with the market and customer’s purchasing behavior. In pursuance of finding more valuable rules, our basic aim is to implement Apriori Algorithm using multithreading approach which can utilization our multicore processing system to improve the performance of an algorithm in a practical and efficient way to unearth more value information for the proper analysis of the business trends.