Pakistan Science Abstracts
Article details & metrics
No Detail Found!!
SVM vs KNN in medical image classification.
Author(s):
1. A. Mueen: Faculty of Information Technology, Multimedia University, Malaysia
2. R. Zainuddin: Faculty of Computer Science & Information Technology, University of Malaya
3. M. Sapiyan Baba: Faculty of Computer Science & Information Technology, University of Malaya
Abstract:
In this paper the authors present comparison results of two most popular classifiers: the Support vector Machine (SVM) and k-Nearest Neighbor (kNN) using medical image database. The classifier kNN is most commonly used image classifier while SVM considered a state of the art classification algorithm. Experiments were performed on IRMA x-ray library dataset, contains 9,000 training x-ray images and 1000 images for testing. The authors concluded with classification results that overall SVM outperforms kNN. In addition, two well suited classification kernel functions, Gaussian Radial Basis function (RBF) and Polynomial (poly) kernel function of SVM are also compared.
Page(s): 164-168
DOI: DOI not available
Published: Journal: Proceedings of International Conference on Information Communication Technologies, Volume: 27, Issue: 0, Year: 2008
Keywords:
Keywords are not available for this article.
References:
References are not available for this document.
Citations
Citations are not available for this document.
0

Citations

0

Downloads

55

Views