Improved CBA classification algorithm based on rough set

Author

Tan, Zheng ; Wang, Hanhu ; Chen, Mei ; Zhang, Xiaoping

Author_Institution

Comput. Sci. & Technol. Dept., Guizhou Univ., Guiyang, China

fYear

2009

fDate

28-31 July 2009

Firstpage

43

Lastpage

46

Abstract

CBA is a classification algorithm integrating association rule mining and classification. CBA has been widely used in data mining areas because it has higher accuracy than C4.5. When the samples become more and more large and characteristic attributes become more and more numerous, CBA algorithm becomes much lower. In this paper, an improved CBA algorithm based on rough set is proposed. The improved CBA algorithm applies rough set to induce attributes, and prune candidate rules with PEP method. Experimental results illustrate that the improved CBA algorithm is efficient and it has higher accuracy than CBA and C4.5.

Keywords

data mining; pattern classification; rough set theory; association rule mining; classification algorithm; data mining; rough set; Association rules; Classification algorithms; Computer science; Data mining; Information systems; Medical diagnosis; Rough sets; Set theory; Symmetric matrices; Text categorization; Attributes induction; CBA classification; Data mining; PEP; Rough set;

fLanguage

English

Publisher

ieee

Conference_Titel

Networked Digital Technologies, 2009. NDT '09. First International Conference on

Conference_Location

Ostrava

Print_ISBN

978-1-4244-4614-8

Electronic_ISBN

978-1-4244-4615-5

Type

conf

DOI

10.1109/NDT.2009.5272128

Filename

5272128