Development of Utility Based Improved Association Rule Mining Algorithm

Abstract

The advent of internet has changed the entire business perspective and most of the business firms primarily depend on the copious digital data available on the net. Mining interesting data related to utilities based on profit and business growth is an ever evolving area and of late utility mining cater to the need of the business houses largely. The most significant tasks in data mining are the process of discovering frequent item sets based on utilities. Incorporating utility considerations in data mining tasks is attaining popularity and admiration recently. It is quite evident that association rules enhance and augment the business value and the data mining groups has approved the mining of these rules of interest. Several business applications have been found to benefit from the discovery of frequent item sets and association rules from transaction databases. A comprehensive survey and study of various methods in existence for association rule mining with utility considerations have been presented and compared in this thesis and a new algorithm named Improved Utility Mining Algorithm , an association rule based utility mining algorithm with superior pruning techniques that restricts the search space, uses very little memory, faster execution speed and less computational costs is proposed and developed in this research work. newline