Multi-class contour preserving classification
by Piyabute Fuangkhon; Thitipong Tanprasert
Title: | Multi-class contour preserving classification |
Author(s): | Piyabute Fuangkhon
Thitipong Tanprasert |
Issued date: | 2012-08 |
Citation: | International Conference on Intelligent Data Engineering and Automated Learning (IDEAL’12), Natal, Brazil, 29-31 August 2012 Lecture Notes in Computer Science LNCS 7435, Springer Berlin / Heidelberg pp. 35-42, ISSN: 0302-9743 (print), ISBN: 978-3-642-32638-7, doi:10.1007/978-3-642-32639-4_5. |
Abstract: |
The original contour preserving classification technique was proposed to improve the robustness and weight fault tolerance of a neu- ral network applied with a two-class linearly separable problem. It was recently found to be improving the level of accuracy of two-class classi- fication. This paper presents an augmentation of the original technique to improve the level of accuracy of multi-class classification by better preservation of the shape or distribution model of a multi-class problem. The test results on six real world multi-class datasets from UCI ma- chine learning repository present that the proposed technique supports multi-class data and can improve the level of accuracy of multi-class classification more effectively. |
Keyword(s): | Contour preserving classification
Data preprocessor Neural networks (Computer science) Outpost vector Pattern classification |
Resource type: | Conference Paper |
Extent: | 8 pages |
Type: | Text |
File type: | application/pdf |
Language: | eng |
Rights holder(s): | Piyabuth Fuangkhon Thitipong Tanprasert |
URI: | http://repository.au.edu/handle/6623004553/20662 |
Files in this item (EXCERPT) |
|
View Conference-Paper-Abstract-20662.PDF ( 204.92 KB ) |
This item appears in the following Collection(s) |
|
|