Research Article
New Fuzzy Support Vector Machine for the Class Imbalance Problem in Medical Datasets Classification
Algorithm 2
Kernel FSVM-CIP.
Input: | training samples | Testing samples | Output: | The predicted labels of data | Procedure: | (1) Choose a kernel function . Compute the Gram matrix . | (2) Compute fuzzy membership using (31) or (32) for the data | (3) Construct data adjacency graph using nearest neighbors and compute the edge weights matrix with examples | (4) Construct local within-class preserving scatter matrix using (24) | (5) Choose parameters (25); and (26) | (6) Compute using (27) and using (28) with a QP Solver | (7) Using decision function (30) with samples , and output the final class labels |
|