Spam Filtering using SVM with different Kernel Functions
Deepak Kumar, Rahul Kumar · International Journal of Computer Applications · 2016
The growing volume of unwanted bulk e-mail (also known as junk-mail or spam) has generated a need for trustworthy antispam filters. Now a day, many Machine learning techniques have been used which are robotically filter the junk e-mail in much unbeaten rate. In this paper, we used one of the most popular machine learning Algorithm support vector machine (SVM) with different parameters using different kernelfunctions (linear, polynomial, RBF, sigmoid) are implemented on spambase-dataset. Comparison of SVM performance for all kernels (linear, polynomial, RBF, sigmoid) using different parameters (C-SVC, NU-SVC) evaluated on spambase-dataset to get best accuracy.