Association rule mining using directed graphs and hypergraphs

Abstract

Data mining is a potential tool for prolific analysis of data The task of association rule mining in a large database is one of the important functionalities in data mining Mining association rules from large databases play an essential role in many data mining tasks and have broad applications Most of the previously proposed methods adopt Apriori like candidate generation and test approaches and FP growth divide and conquer approach to decompose both the mining task and the databases However these methods may encounter serious challenges when mining frequent patterns for generating association rules newlineIn this work we develop a directed graph model along with a path system for representing a large transaction database This model is used for mining frequent patterns from large databases of transactions and for generating strong association rules Directed graph model adopts a divide and conquer and a greedy approach to decompose the mining task and the database Then path system in directed graphs is used to avoid the costly candidate generation and test processing completely Moreover, effective data structures are proposed to compress crucial information about frequent patterns and avoid expensive repeated database scan newline newline