Realtime Spam Detection System Using Naive Bayes Algorithm in Comparison With Support Vector Machine Learning Algorithm

The aim of this study is to predict spam messages using Naive Bayes in comparison with SVM algorithms to improve the accuracy. Spam message prediction is performed using support vector machine algorithm (N=150) over Naive Bayes machine learning algorithm (N=150) with the split size of training and testing dataset 70% and 30% respectively. The retrieval accuracy of SVM classifier is (93%) and Naive Bayes algorithm is (90%) and attained significance value of p = 0.01. The work has confirmed that the efficiency of the support vector machine algorithm has given more accuracy in detecting spam messages when compared to Naive Bayes machine learning algorithm.