Performance comparison of support vector machine and gaussian naive bayes classifier for youtube spam comment detection

Abstract

Youtube is a video sharing site that was begun back in 2005. Youtube produces over 400 hours of substance each moment and more than 1 billion hours of substance are devoured by clients every day. In this work, we present a new approach by comparing the analysis results using a support vector machine and the Gaussian Naive Bayes classificatio. Our proposed methodology We used the  dataset from UCI especially Youtube-Shakira for training and testing. The transformed dataset is split into training and testing subsets and fed into Naive Bayes and Support Vector Machin. In all cases, the F1 score was used to evaluate the classifier's performance. The results of the experiment are displayed in Gaussian Naive Bayes with an F1 score of 84.38% and a Support Vector Machine (SVM) with an F1 score of 88.00%. Naive Bayes is consistently the worst performer than SVM.