• Linear Kernel
• Polynomial Kernel
• Radial Basis Function (RBF)
• Laplacian Kernel
• Sigmoid Kernel are implemented without using any inbuilt library and studied the importance of using and not using these kernel in the applications like Classification using Support Vector Machine and Principle Component Analysis.

A data containing Non-linear separable data points is taken and studied on which kernel can better project the above data into higher dimensional while using SVM as the classification algorithm so that it can linearly separable.

Here the separation of the data points before and after applying kernel are plotted using 3d scatter plot available from the plotly library

In a similar way to the first application, non-linearly separable data is taken and studied how the PCA and Kernel functions applied PCA can project the data points into higher dimensions and make them linearly separable.