NEWSIFY- Categorisation and Perception of Suspicious News using Machine Learning Algorithm- A Survey

Reading news articles is an integral part of our daily life. Journalists and editors are tasked with determining which articles are popular so they can efficiently allocate resources to support a better reading experience. This paper aims to achieve a model which simultaneously performs all the three aspects i.e: detecting whether the given news article is fake or not, summarizing a given news article and predicting the popularity and ranking the news article. The reasons behind the popularity of news articles tend to be varied and may include contemporaneity, writing quality, and other potential factors. In this article, we consider the problem of popularity prediction as a regression, developing several classes of features (metadata, contextual or content-based, temporal and social) and building models to predict popularity. The system presented here will be used in a real time environment. The other aspect of the paper is a summary of news articles on a particular topic. Since there are also various online resources that publish fake news, it is difficult to judge and understand which articles are fake. This paper focuses on the classification of the news as fake or real. The different Natural Language Processing and Machine Learning algorithms are studied for the implementation of the model. This paper helps the users for the identification of real and fake news which are published online. This will satisfy the problems of users willing to obtain quality news content in a short period of time.