Accession number:20090111824561
Title:A choquet integral regression model with a new fuzzy measure based on multiple mutual-information
Authors:Liu, Hsiang-Chuan (1); Chang, Horng-Jinh (2); Lin, Wen-Chih (3); Chang, Kai-Y.I. (4)
Author affiliation:(1) Department of Bioinformatics, Asia University, Taiwan; (2) Department of Psychology, Asia University, Taiwan; (3) Department of Computer Science and Information Engineering, Asia University, Taiwan; (4) Graduate Institute of Educational
Measurement and Statistics, Taichung University, Taiwan Corresponding author:Liu, H.-C.
Source title:Proceedings of the 7th International Conference on Machine Learning and Cybernetics, ICMLC
Abbreviated source title:Proc. Int. Conf. Mach. Learn. Cybern., ICMLC Volume:6
Monograph title:Proceedings of the 7th International Conference on Machine Learning and Cybernetics, ICMLC
Issue date:2008
Publication year:2008 Pages:3558-3562
Article number:4621021 Language:English
ISBN-13:9781424420964
Document type:Conference article (CA)
Conference name:7th International Conference on Machine Learning and Cybernetics, ICMLC
Conference date:July 12, 2008 - July 15, 2008 Conference location:Kunming, China
Conference code:74802
Publisher:Inst. of Elec. and Elec. Eng. Computer Society, 445 Hoes Lane - P.O.Box 1331, Piscataway, NJ 08855-1331, United States Abstract:The well known fuzzy measures, λ -measure has no information with the dependent variable. Owing to above problem, the Ε -measure based on multiple entropy is proposed by our previous study. In this paper, an improved fuzzy measure based on multiple mutual-information, called M-measure, is
proposed. For evaluating the Choquet integral regression models with different fuzzy measures, a real data experiment by using a 5- fold cross validation mean square error (MSE) is conducted. The performances of the Choquet integral regression models based on M-measure, Ε -measure and λ -measure,
respectively, a ridge regression model, and the traditional multiple linear regression model are compared. Experimental result shows that Choquet integral regression model based on the new measure, M-measure, has the best performance. © 2008 IEEE.
Number of references:7
Main heading:Regression analysis
Controlled terms:Control theory - Cybernetics - Integral equations - Learning systems - Linear regression - Mean square error - Robot learning
Uncontrolled terms:Choquet integral regression model - Cross validations - Dependent variables - Fuzzy measures - M-measure - Mean squares - Multiple linear regression models - Multiple mutual- information - Real datums - Ridge regressions
Classification code:921.2 Calculus - 731.5 Robotics - 731.1 Control Systems - 922.2 Mathematical Statistics - 723.5 Computer
Applications - 461.9 Biology - 461.4 Ergonomics and Human Factors Engineering - 723.4 Artificial Intelligence
DOI:10.1109/ICMLC.2008.4621021 Database:Compendex
Compilation and indexing terms, Copyright 2009 Elsevier Inc.
