Intelligent Agents for Data Mining and Information Retrieval [Electronic resources] نسخه متنی

Chapter XVII: An Algorithm of Pattern Match Being Fit for Mining Association Rules

Hong Shi,

Taiyuan Heavy Machinery Institute, China

Ji-Fu Zhang,

Beijing Institute of Technology, China

There are frequent occurrences of pattern match involved in the process of counting the support count of candidates, which is one of the main factors influencing the efficiency of mining for association rules. In this chapter, an efficient algorithm for pattern match being fit for mining association rules is presented by analyzing its characters, and it has been proved correctly and efficiently.

PRODUCTION

Association rules are one of the knowledge models in data mining. R. Agrawal developed the concept of association rules, which implies the relationships among a set of objects of transaction data set DB. Efficiency is the key to mining algorithms, owing to the frequent amounts of data included in DB. At present, the most effective algorithm of mining association rules is Apriori algorithm, presented by Agrawal and Srikant (1994).

Mining association rules may require iterative scanning of frequent transaction data sets and matching with candidates to count the support. Owing to the frequent data sets, the process of match is the main factor in efficiency. To resolve the question, we present a highly efficient pattern match algorithm being fit for mining association rules during exploiting, and we research the Market Basket Data Analysis System Based on Mining Associations Rules by analyzing some of the characteristics of pattern match included in mining association rules.