Choquet integral regression model based on high-order L-measure

The well known fuzzy measures, lambda-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...

Full description

Saved in:
Bibliographic Details
Main Authors: Hsiang-Chuan Liu, Wei-Sung Chen, Yu-Chieh Tu, Yen-Kuei Yu
Format: Conference Proceeding
Language:English
Subjects:
Asia
Bioinformatics
Choquet integral regression model
Computer science
Cybernetics
Density measurement
Fuzzy sets
L h -measure
L-measure
Linear regression
Machine learning
P-measure
Predictive models
Statistics
λ-measure
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The well known fuzzy measures, lambda-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 ldquohth-order L-measurerdquo, denoted as L h -measure, and a new Choquet integral regression model based on this L h -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 lambda-measure, P-measure, L-measure and L h -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 h -measure based on gamma-support outperforms others forecasting models.
ISSN:2160-133X
DOI:10.1109/ICMLC.2009.5212800