SVM with multiple kernels based on manifold learning for Breast Cancer diagnosis

In this paper, we propose an efficient algorithm Support Vector Machines with multiple kernels based on Isometric feature mapping(Isomap) in the process of breast cancer classification. We use Wisconsin Diagnostic Breast Cancer (WDBC) as our original data set. The first step, we use Isomap to project high dimensional breast cancer data into a much lower dimensional space. Second, we use SVM with multiple kernels to classify the lower dimensional breast cancer data. Finally, the experimental results illustrate that the proposed algorithm has a better performance than traditional SVM for breast cancer classification.