Improved RElim and FIN algorithm for frequent items generation

Abstract

Data mining is a process of extracting hidden information from large databases. Data mining is basically focused on many areas like — communication, retail, Financial, and marketing organizations. It determines relationships among internal and external factors. Association rule is a method for identifying the relations between variables in large databases. It is determined to discover frequent patterns, identify rules and strong rules discovered in databases. The main objective of this research work is to find accurate and large number of frequent itemset by enhancing existing algorithms like FIN and RElim algorithm, frequent patterns are generated and strong rules are identified. Normally in Association rule mining common threshold value is given to find the frequent itemset but in the enhanced algorithms individual threshold values are given to every item in the transactional database to find out the frequent items. From the analysis it was observed that enhanced RElim algorithm gives best results than enhanced FIN algorithm.