Abstract:
In recent years sharing information through internet and various social platform has increased rapidly. It is very hard to find the credibility of those information. People are sharing news without knowing the proper fact. Considering the harmful effect that can be caused by this, detection of fake news has become a new phenomenon of the society. The research process of fake news detection is still in early stage. This paper aims to detect fake news with the help of Natural Language processing (NLP). For this proposed method, three different classifier- logistic regression, decision tree and random forest have been used. The labeled dataset has been collected from a public domain. Use of TF-IDF vectorizer has been made. In this paper the challenge, task formulation and all the steps of the NLP solution has been discussed. The comparison among the classifiers is also shown in this paper.
