Textual Analysis and Identification of Spam News

Abstract

Abstract: Our study aims to tackle the problem of recognizing spam news through the use of Machine Learning. To accomplish this, we devised a hybrid approach named "SpamBuster" that employs various algorithms including Bernoulli Naive Bayes, Multinomial Naive Bayes, Random Forest, and Kernel SVM. We trained the model on a large dataset of news articles and evaluated it based on multiple performance metrics such as accuracy, precision, recall, and F1 score. Our experimental findings revealed that our approach achieved outstanding performance with an accuracy of 0.9507, precision of 0.9747, recall of 0.9253, and F1 score of 0.9494. These results demonstrate that our approach is effective in identifying spam news and could have practical implications in combating the spread of false information and propaganda. In conclusion, our research showcases the potential of machine learning techniques and textual analysis for detecting spam news and emphasizes the significance of this area of research in the modern era of information.