Can I read An Incremental Approach to Contribution-Based Feature Selection on EtoBox?

An Incremental Approach to Contribution-Based Feature Selection by Sheng-Uei Guan; Jun Liu; Yinan Qi is a Computer Science article available to read on EtoBox.

What is An Incremental Approach to Contribution-Based Feature Selection about?

This paper presents a novel feature selection approach based on an incremental neural network (NN) training approach. Instead of training input attributes in batch, this incremental approach trains input attributes one by one so that network performance keeps refined when each new attribute comes in. If an incoming attribute is consistent with previous attributes and relevant to output attributes, network performance will be improved, otherwise degraded. The contribution of an input attribute is evaluated through network performance evaluation. Attributes with little or no contribution will be discarded. To have fair feature selection, we evaluate the individual discrimination ability of each attribute before training by using a NN with only one attribute in the input layer. The attribute with the best discrimination ability will be introduced first, followed by those attributes with lower discrimination ability. Two feature-detection methods are discussed based on this incremental training approach. Unlike existing feature selection methods, the proposed feature selection methods are suitable not only for classification problems but also for regression problems. Experimental resul

Who reads An Incremental Approach to Contribution-Based Feature Selection?

Author

Sheng-Uei Guan; Jun Liu; Yinan Qi

Publisher

Walter de Gruyter GmbH

Published

2004

Language

EN

Field

Computer Science (Physical Sciences)