Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/56035
Title: Feature selection for modular neural network classifiers
Authors: Guan, S.-U. 
Li, P. 
Keywords: Class decomposition
Correlation between input features
Feature selection
FLD
Modular neural network
Transformation vector
Issue Date: 2002
Source: Guan, S.-U.,Li, P. (2002). Feature selection for modular neural network classifiers. Journal of Intelligent Systems 12 (3) : 173-200. ScholarBank@NUS Repository.
Abstract: An N-class problem can be fully decomposed into N-independent small neural networks called modules (or sub-problems) in a modular neural network classifier. Each sub-problem is a two-class ('yes' or 'no') problem. Hence, the optimal input feature space for each module is also likely to be a subset of the original feature space. Therefore, feature selection plays an important role in finding these useful features. Some feature selection techniques have been developed from different perspectives but are not suitable, however, for the two-class problems resulting from complete task decomposition. In this paper, we propose two feature selection techniques- Relative Importance Factor (RIF) and Relative FLD Weight Analysis (RFWA) for modular neural network classifiers. Our approaches involve the use of Fisher's linear discriminant (FLD) function to obtain the importance of each feature and to find the correlation among features. In RIF, the input features are classified as relevant and irrelevant based on their contribution in classification. In RFWA, the irrelevant features are further classified into noise or redundant features based on the correlation among features. The proposed techniques have been applied to several classification problems. The results show that these techniques can successfully detect the irrelevant features in each module and improve accuracy while reducing computation effort.
Source Title: Journal of Intelligent Systems
URI: http://scholarbank.nus.edu.sg/handle/10635/56035
ISSN: 03341860
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

Page view(s)

21
checked on Dec 8, 2017

Google ScholarTM

Check


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