Detection of breast cancer using v-SVM and RBF networks with self organized selection of centers
Tingting Mu · 2005
In this paper we propose, for the first time, to apply v-SVM learning instead of the original and commonly used c-SVM learning to breast cancer detection, and perform v-SVM parameter selection based on the restricted leave-one-out error estimate using grid search with no need for validation data. An efficient method of radial basis function networks based on the self-organizing clustering results has also been applied to improve the detection performance of using only self-organizing maps. Wisconsin diagnosis breast cancer dataset is used to evaluate our proposed methods. Experimental results demonstrate that our proposed methods offer better performance compared with other existing methods.