Affiliations: State Key Lab. of Industrial Control Technology,
Zhejiang University, China Institute of Advanced Process Control, Zhejiang
University, China. E-mail: yongliu@iipc.zju.edu.cn | College of Computer Science, Zhejiang University,
China
Note: [] Address for correspondence: State Key Lab. of Industrial Control
Technology, Zhejiang University, 310027 Hangzhou, China
Abstract: Rough rule extraction refers to the rule induction method by using
rough set theory. Although rough set theory is a powerful mathematical tool in
dealing with vagueness and uncertainty in data sets, it is lack of effective
rule extracting approach under complex conditions. This paper proposes several
algorithms to perform rough rule extraction from data sets with different
properties. Firstly, in order to obtain uncertainty rules from inconsistent
data, we introduce the concept of confidence factor into the rule extracting
process. Then, an improved incremental rule extracting algorithm is proposed
based on the analysis of the incremental data categories. Finally, above
algorithms are further extended to perform approximate rule extraction from
huge data sets. Preliminary experiment results are encouraging.
