Developer Guide for Intel® Data Analytics Acceleration Library 2016 Update 4
A linear kernel is the simplest kernel function.
Given a set X of n feature vectors x1= (x11,…,x1p), ..., xn= (xn1,…,xnp) of dimension p and a set Y of m feature vectors y1= (y11,…,y1p), ..., yn= (ym1,…,ymp), the problem is to compute the linear kernel function K (xi, yi) for any pair of input vectors: K (xi, yi) = k xi Tyi + b.