Abstract
Fake news detection is an evolving area of research nowadays. This area involves quite a lot of research due to inadequacy of available resources. The problem of fake news is causing a detrimental effect on society. Because of bad societal effects due to false information, its detection has attracted increasing attention. We have presented a Fake News Detection Tool (FNDT) using various Natural Language Processing and Machine Learning techniques. Our proposed tool is based on feature selection approaches: Bag of Words and TF-IDF. We have investigated and compared the performance of different classification algorithms using these approaches. For implementation purposes, we have taken a standard LIAR dataset. The results have shown that the Random Forest classifier results out to be most fitting and has an F1 score metric value of 0.703 by using the Bag of Words approach. The Naïve Bayes classifier has performed the best and has an F1 score metric value of 0.723 by using the TF-IDF approach.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call