Scheduled System Maintenance:
On May 6th, single article purchases and IEEE account management will be unavailable from 8:00 AM - 12:00 PM ET (12:00 - 16:00 UTC). We apologize for the inconvenience.
By Topic

Static and Dynamic Weights in Ensemble Systems Built by Class-Based Feature Selection Methods

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$31 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Vale, K.O. ; Dept. of Inf. & Appl. Math., Fed. Univ. of RN, Natal, Brazil ; Neto, A.F. ; Canuto, A.M.P. ; Dias, F.G.

The use of feature selection methods in ensemble systems has been shown to be efficient, since it reduces the dimensionality while increases the diversity among the individual classifiers of these systems. The ReinSel method, a simple reinforcement-based process, for instance, has been proposed to select feature for the individual classifiers of an ensemble system. This method distributes the attributes through the use of a class-based process (using One-Against-All, OAA, classifiers). In this paper, we investigate the use of weights in order to enhance the efficiency of the ensemble systems created by class-based feature selection methods. These weights will not be used in feature selection methods, but in the ensemble systems created as the result of these methods. More specifically, four different types of weights will be used in this investigation, in which three of them are defined before the testing phase and became unchanged during the testing phase (static). The last one uses a knn-based method to define the weights for each testing pattern (dynamic).

Published in:

Neural Networks (SBRN), 2010 Eleventh Brazilian Symposium on

Date of Conference:

23-28 Oct. 2010