Fake News Detection Using Machine Learning and Deep Learning Algorithms

Classification of fake news on social media has gained a lot of attention in the last decade due to the ease of adding fake content through social media sites. In addition, people prefer to get news on social media instead of on traditional televisions. These trends have led to an increased interest in fake news and its identification by researchers. This study focused on classifying fake news on social media with textual content (text classification). In this classification, four traditional methods were applied to extract features from texts (term frequency-inverse document frequency, count vector, character level vector, and N-Gram level vector), employing 10 different machine learning and deep learning classifiers to categorize the fake news dataset. The results obtained showed that fake news with textual content can indeed be classified, especially using a convolutional neural network. This study obtained an accuracy range of 81 to 100% using different classifiers.