Fake news Detection Using Naive Bayes Classifier

Abstract

Fake news has been on the rise thanks to rapid digitalization across all platforms and mediums. Many governments throughout the world are attempting to address this issue. The use of Natural Language Processing and Machine Learning techniques to properly identify fake news is the subject of this research. The data is cleaned, and feature extraction is performed using pre-processing techniques. Then, employing four distinct strategies, a false news detection model is created. Finally, the research examines and contrasts the accuracy of Naive Bayes, Support Vector Machine (SVM), neural network, and long short-term memory (LSTM) methodologies in order to determine which is the most accurate. To clean the data and conduct feature extraction, pre-processing technologies are needed. Then, employing four distinct strategies, a false news detection model is created. Finally, in order to determine the best fit for the model, the research explores and analyzes the accuracy of Naive Bayes, Support Vector Machine (SVM), neural network, and long short-term memory (LSTM) approaches. The proposed model is working well with an accuracy of products up to 93.6%.