Table 2 The four basic kernel functions
From: Exploring the potential of 3D Zernike descriptors and SVM for protein–protein interface prediction
Kernel name | Mathematical formulation |
|---|---|
Linear | \(K(\boldsymbol {x}_{i}, \boldsymbol {x}_{j})=\boldsymbol {x}_{i}^{\intercal } \boldsymbol {x}_{j}\) |
Polynomial | \(K(\boldsymbol {x}_{i}, \boldsymbol {x}_{j})=\left (\gamma \boldsymbol {x}_{i}^{\intercal } \boldsymbol {x}_{j} + r\right)^{d}, \gamma > 0\) |
Radial basis function (RBF) | K(x i ,x j )=exp(−γ∥x i −x j ∥2),γ>0 |
Sigmoid | \(K(\boldsymbol {x}_{i}, \boldsymbol {x}_{j})=\text {tanh}\left (\gamma \boldsymbol {x}_{i}^{\intercal } \boldsymbol {x}_{j} + r\right), \gamma > 0\) |