| Paper Details: | Downloads: 254 |
| Serial Number: | P1121140859
|
| Title: | Wavelet Neural Networks with a New Translation Vector Initialization Approach for Multiclass Cancer Classification
|
| Authors: | Zarita Zainuddin, Ong Pauline
|
| Abstract: | The generalization capability of wavelet neural networks (WNNs) is sensitively influenced by the network topology, particularly the location and the number of translation vectors. In this paper, a new translation vector initialization approach – specifically, the modified point symmetry-based fuzzy C-means (MPSDFCM) was employed in selecting the translation vectors of WNNs, where its effectiveness was demonstrated in the multiclass cancer classification using the microarray gene expression profiles. Comparative studies showed that the proposed methodology possessed superior predictive capability as opposed to other initialization approaches, in terms of prediction accuracy, sensitivity and specificity.
|
| Keywords: | Bioinformatics, Fuzzy C-Means, Microarray, Wavelet Neural Networks
|
| Journal/Conference: | ICGST Conference on Computer Science and Engineering, CSE-11
|
| Volume: |
|
| Issue: |
|
| Submission Date: | 10/2/2011 12:00:00 AM
|
| Review Date: | 11/29/2011 7:12:53 PM
|
| Publishing Date: | 12/19/2011 12:00:00 AM
|
| Article Downloads: | 254
|
| Download: |
|