Many email services today provide spam filters that are able to classify emails into spam and non-spam email with high accuracy.The project is to develop a support vector machine using a Gaussian kernel capable of performing such a task.
The project was done as a part of Stanford ML Course on Coursera.
The project consists of 3 datasets .The first dataset uses basic linear decision boundary where the C value is constant. The second dataset involves using a Gaussian kernel which is based on linearly inseparable dataset.The 3rd dataset is more advance once requiring the use of both.
For further reading please refer to this pdf