A Fast Algorithm of Mining Multidimensional Association Rules Frequently

Author

Xu, Wan-xin ; Wang, Ru-Jing

Author_Institution

Hefei Inst. of Intelligent Machines, Chinese Acad. of Sci., Hefei

fYear

2006

fDate

13-16 Aug. 2006

Firstpage

1199

Lastpage

1203

Abstract

In this paper, a novel algorithm named MDFM is proposed. It is an algorithm for mining multidimensional association rules frequently from relational database. The algorithm uses a structure including many indexes and bases on statistic idea. When using the algorithm for the first time, it scans the target database only once and all frequent itemsets and association rules can be generated. After some parameters are adjusted, it is not necessary to scan the database at all and all frequent itemsets can be generated when the algorithm runs again. So, it can be used when the target database must be mined time after time. Compared with some traditional algorithms of mining association rules, the algorithm presented in this paper has better executive efficiency and expansibility, which is proved in our experiments

Keywords

data mining; data structures; database indexing; relational databases; statistical analysis; data mining; data structure; database index; frequent itemset; multidimensional association rule; relational database; statistic analysis; Association rules; Automation; Costs; Cybernetics; Data mining; Itemsets; Machine intelligence; Machine learning; Machine learning algorithms; Multidimensional systems; Relational databases; Transaction databases; Data mining; frequent itemset; index; multidimensional association rule; relational database;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2006 International Conference on

Conference_Location

Dalian, China

Print_ISBN

1-4244-0061-9

Type

conf

DOI

10.1109/ICMLC.2006.258605

Filename

4028246