Implication of association rules employing FP-growth algorithm for knowledge discovery

Abstract

Nowadays the database of an organization is increasing day by day. Sometimes it is necessary to know the behavior of that organization by retrieving the relationships among different attributes of their database. Implication of association rules provides an efficient way of data mining task which is used to find out the relationships among the items or the attributes of a database. This paper addresses on implication of association rules among the quantitative and categorical attributes of a database employing classical logic and Frequent Pattern (FP) - Growth algorithm. The system is based on generating association rules over binary or categorical attributes and is organized with splitting the quantitative attributes into two or more intervals to generate association rules when the domain of quantitative attribute increases. The effectiveness of the method has been justified over a sample database.