Research Article
Privacy Preserving RBF Kernel Support Vector Machine
Algorithm 1
Hybrid SVM algorithm.
Input: Public data , private data , the dimensionality of , a regularization parameter , | and privacy budget ; | Output: Differentially private SVM; | (1) Use the public data to compute via (5), (6); | (2) Transform each record of the private data to new 2D-dimensional data via the mapping function | defined by (7); | (3) Compute the parameter in the dual space with the transformed private data, and in the primal space | via ; | (4) Draw from Lap, , then return and . |
|