DocumentCode
477744
Title
The Sugeno Integral and Its Applications of Nonadditive Set Functions Based on Sigma-Lambda Rules
Author
Chen, Li ; Gong, Zengtai ; Duan, Gang
Author_Institution
Dept. of Math., Lanzhou City Univ., Lanzhou
Volume
1
fYear
2008
fDate
18-20 Oct. 2008
Firstpage
650
Lastpage
654
Abstract
Sugeno integral is used as an aggregation tool in information fusion which could be regarded as the inverse problem of information fusion as well as in nonlinear multi-regressions and nonlinear classification. Introducing more types of nonlinear integrals with provide more choices for various of objectives and components in these problems. However, when sample space is huge, regarding needs the information contribution to obtain is very difficult, this paper has carried on the discussion on this. In this paper, first we introduce a kind of nonadditive measure based on sigma-lambda rules and its calculate of the defect of additivity are given. Then based on this, we could calculate contributions of their joint attributes by means of sigma-lambda rules based on the contributions of individual attributes. Finally, the lower integral and upper Sugeno integral of the nonadditive set functions based on sigma-lambda rules are defined and calculated by the integer program problems.
Keywords
integer programming; integral equations; inverse problems; nonlinear equations; pattern classification; regression analysis; set theory; Sugeno integral; aggregation tool; information fusion; integer program problems; inverse problem; nonadditive set functions; nonlinear classification; nonlinear integrals; nonlinear multiregressions; sigma-lambda rules; Databases; Educational institutions; Fuzzy systems; Information science; Inverse problems; Mathematical model; Mathematics; Transportation;
fLanguage
English
Publisher
ieee
Conference_Titel
Fuzzy Systems and Knowledge Discovery, 2008. FSKD '08. Fifth International Conference on
Conference_Location
Shandong
Print_ISBN
978-0-7695-3305-6
Type
conf
DOI
10.1109/FSKD.2008.668
Filename
4666056
Link To Document