Mining Association Rule by ECLAT Method Using Transaction Data

Abstract

Association rule mining is a process that identifies links between sets of correlated objects in transactional databases where each transaction contains a list of items. Association rule is one of the well-defined algorithms, whose significance is measured via support and confidence factor, are intended to identify rules of the type. This system is the development of transactions data analysis system. The important problems of data mining are mining frequent itemsets and generating association rules from databases of transactions where each transaction consists of a set of items. Our proposed system is based on Association Rule Mining using Equivalence CLASS Transformation (ECLAT) method to find frequent-patterns. This method can also reduce the number of candidate itemsets. It is not required scanning the complete database over and over again. So, it also saves the time.