Fake News Detection: An Ensemble Learning Approach

Due to easy access, rapid growth, and proliferation of the information available through regular news mediums or social media, it is becoming easy for people to look for news and consume it. But on the other hand, it is becoming a daunting task to differentiate between false information and true information thus leading to widespread fake news. We can define fake news as a type of deceiving journalism and statements that are used to artifice and mislead people. Also, the credibility of social media platforms is at stake where this news is mostly shared. These types of forged information news can have serious negative societal impacts and thus their detection has become the emerging area that is attracting research attention. In this paper, we propose a model for detecting fake news by examining the accuracy of a report and predicting its authenticity. By feature extraction and forming credibility scores from the textual information, this model builds an ensemble network to learn the portrayals of news reports, authors, and titles simultaneously. Different machine learning algorithms like SVM, CNN, LSTM, KNN, and Naive Bayes are used for higher accuracy and it was observed that LSTM showed better accuracy with 97%. The performance and effectiveness of classifiers were evaluated based on their precision, recall, and F1-Score. The usage of different algorithms shows the effectiveness of performance on the dataset.