Fuzzy partial correlation rules mining

Author

Chueh, Hao-en ; Lin, Nancy P.

Author_Institution

Dept. of Inf. Manage., Yuanpei Universit, Hsinchu, Taiwan

Volume

1

fYear

2009

fDate

28-29 Nov. 2009

Firstpage

91

Lastpage

94

Abstract

Mining fuzzy correlation rules is the task of finding correlation relationship between the two fuzzy itemsets in a transactional database. In many situations, however, fuzzy itemsets other than the two under consideration are also responsible for the observed correlation relationship, the effects of these fuzzy itemsets may influence the observed correlation relationship, and thus, the real correlation relationship cannot be really obtained. Therefore, in this research, the analysis of fuzzy partial correlation is used to construct a new algorithm for discovering the fuzzy partial correlation rule. The fuzzy partial correlation analysis can provide us the correlation relationship between two fuzzy itemsets when the influences of other fuzzy itemsets are held constant. Thus, by using the fuzzy partial correlation analysis, the fuzzy partial correlation rules which can show us the strong correlation relationship between the fuzzy itemsets when the influences of other fuzzy itemsets are removed can be effectively generated.

Keywords

correlation methods; data mining; fuzzy set theory; fuzzy data mining; fuzzy itemsets; fuzzy partial correlation analysis; fuzzy partial correlation rules mining; Application software; Association rules; Computational intelligence; Computer industry; Data mining; Fuzzy sets; Industrial relations; Itemsets; Mining industry; Transaction databases; fuzzy association rule; fuzzy correlation rule; fuzzy data mining; fuzzy partial correlation rule;

fLanguage

English

Publisher

ieee

Conference_Titel

Computational Intelligence and Industrial Applications, 2009. PACIIA 2009. Asia-Pacific Conference on

Conference_Location

Wuhan

Print_ISBN

978-1-4244-4606-3

Type

conf

DOI

10.1109/PACIIA.2009.5406368

Filename

5406368