Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/21511
Title: Theoretical choice of the optimal threshold for possibilistic linear model with noisy input
Authors: Ge, H
Chung, FL 
Wang, S
Keywords: Maximum a posteriori (MAP)
Possibilistic linear model (PLM)
Possibility theory
Issue Date: 2008
Publisher: Institute of Electrical and Electronics Engineers
Source: IEEE transactions on fuzzy systems, 2008, v. 16, no. 4, p. 1027-1037 How to cite?
Journal: IEEE transactions on fuzzy systems 
Abstract: Based on possibility concepts, various possibilistic linear models (PLMs) have been proposed, and their pivotal role in fuzzy modeling and associated applications has been established. When adopting PLMs, one has to adopt an appropriate threshold (λ) value. However, choosing such a value is by no means trivial, and is still an open theoretical issue. In this paper, we propose a solution by first extending the PLM to its regularized version, i.e., a regularized PLM (RPLM), such that its generalization capability can be enhanced. The RPLM is then formulated as a maximum a posteriori (MAP) framework, which facilitates the determination of the theoretically optimal threshold value for the RPLM with noisy input. Our mathematical derivations reveal the approximately inversely proportional relationship between the threshold λ and the standard deviation of Gaussian noisy input. This is also confirmed by the simulation results. This finding is very helpful for the practical applications of both PLMs and RPLMs.
URI: http://hdl.handle.net/10397/21511
ISSN: 1063-6706
EISSN: 1941-0034
DOI: 10.1109/TFUZZ.2008.917290
Appears in Collections:Journal/Magazine Article

SFX Query Show full item record

SCOPUSTM   
Citations

4
Last Week
0
Last month
0
Citations as of Dec 10, 2017

WEB OF SCIENCETM
Citations

1
Last Week
0
Last month
0
Citations as of Dec 18, 2017

Page view(s)

33
Last Week
1
Last month
Citations as of Dec 18, 2017

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.