Differential Privacy Protection for Support Vector Machines for Nonlinear Classification

Table 1

Summary of main notations.

Notations	Meaning
D	A set of samples
	Feature of ith sample
	Label of ith sample
n	The number of samples
d	The number of attributes
	The privacy budget
	Proportional parameter
	Global sensitivity of the training data set
	Training data set after adding noise
K	Kernel function
	Global sensitivity of the kernel function
	Kernel function after adding noise
γ	Parameter of Gaussian kernel and sigmoid kernel
c	Independent item
N	Laplace noise matrix