Title: Choquet integral regression model based on high-order l- measure
Authors: Liu, Hsiang-Chuan (1); Chen, Wei-Sung (2); Tu, Yu-Chieh (3);
Yu, Yen-Kuei (3)
Author affiliation:(1) Department of Bioinformatics, Asia University, Wufeng, Taiwan; (2) Department of Computer Science and
Information Engineering, Asia University, Wufeng, Taiwan; (3) Graduate Institute of Educational Measurement and Statistics, Taichung University, Taichung, Taiwan
Corresponding author:Liu, H.-C.
Source title: Proceedings of the 2009 International Conference on Machine Learning and Cybernetics
Abbreviated source title:Proc. Int. Conf. Mach. Learn. Cybern.
Volume:6
Monograph title:Proceedings of the 2009 International Conference on Machine Learning and Cybernetics
Issue date:2009
Publication year:2009 Pages:3177-3182
Article number:5212800 Language:English
ISBN-13:9781424437030
Document type:Conference article (CA)
Conference name:2009 International Conference on Machine Learning and Cybernetics
Conference date:July 12, 2009 - July 15, 2009 Conference location:Baoding, China
Conference code:78063
Publisher:IEEE Computer Society, 445 Hoes Lane - P.O.Box 1331, Piscataway, NJ 08855-1331, United States
Abstract:The well known fuzzy measures, λ-measure and P- measure, have only one formulaic solution, the former is not a closed form, and the later is not sensitive. An improved multivalent fuzzy measure with infinitely many solutions of closed form, called L-measure, is proposed by our previous work. In this paper, expend the L-measure for being more choice, and get an improved fuzzy
measures, called "hth-order L-measure", denoted as L
<sup>h</sup>-measure , and a new Choquet integral regression model based on this L<sup>h</sup>-measure is also proposed. For evaluating the proposed 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
Choquet integral regression models with fuzzy measure based on ?- measure, P-measure, L-measure and L<sup>h</sup>-measure, respectively, a ridge regression model, and a multiple linear
regression model are compared. Experimental result shows that the Choquet integral regression models with L<sup>h</sup>-measure based on support outperforms others forecasting models. ©
2009 IEEE.
Number of references:13
Main heading:Integral equations
Controlled terms: Control theory - Cybernetics - Linear regression - Mean square error - Robot learning
Uncontrolled terms: Choquet integral - Choquet integral regression model - Closed form - Cross validation - Forecasting models - Fuzzy measures - High-order - L-measure - L-measure Lh-measure -
Measure P-measure - Multiple linear regression models - Regression model - Ridge regression
Classification code:461.9 Biology - 723.4 Artificial Intelligence - 731.1 Control Systems - 731.5 Robotics - 921.2 Calculus - 922.2 Mathematical Statistics
DOI:10.1109/ICMLC.2009.5212800 Database:Compendex
Compilation and indexing terms, Copyright 2009 Elsevier Inc.
