Improving fake news detection using Deep learning techniques

Abstract

Social media can provide instant news faster than conventional news outlets or sources. It is a great wealth of information, yet there is a growing need to verify this information’s accuracy and correctness. The rate of producing digital information is large and quick, running daily at every second. The developing fame of social media has brought the massive creation of usergenerated content. A significant part of this data is valuable and has turned out to be a great learning source. The growing fame of social media outlets has estimated the distribution of news articles that have caused the fake news explosion. Fake news is made and published with the intent to mislead and damage the representation of an agency, entity, person for commercial and political advantages. With the rapid emergence of fake news, serious attentiveness has produced in our society due to immense fake content distribution. The widespread of fake news has the potential for incredibly adverse effects on individuals and civilization. In this regard, fake news identification via social media platforms has, as of late, turned into an emerging research topic that is drawing huge consideration. Currently, there is no direct method to distinguish whether the information presented as a piece of news is both trustworthy and beneficial. Search engines are the doors to learning, but seeking significance cannot ensure that the matter is reliable. An easygoing observer probably won’t have the capacity to differentiate between reliable and untrustworthy news. My research work is centered on evaluating such imparted news articles on social media for their reliability and trustworthiness. Fundamental theories of trust have utilized to motivate the search for a better solution.