Fake News Classification Using Vectorized Semantic and Syntactical Analysis

Abstract

AbstractThe sudden growth in wireless communication, computing, and insightful device management has led to the rapid spread of the Internet all over the globe. Internet applications and services can be accessed by people at any given time. This rapid growth has made the quality of living better by saving our efforts and time. However, the spread of the Internet has also increased its misuse in online platforms. One example is the extensive spread of fake news over the globe in social, political, and economic contexts. Fake news is news that is deliberately made to deceive the readers. Fake agendas are distributed as real information as news to the readers. Detection of fake news is a bold task for the already present content analysis of traditional models. Lately, feature extraction in neural network models has gotten an edge over the traditional models in detecting fake news. However, there is still a lot of research scope in the field of fake news detection. In this paper, fake news detection is done on news articles that are spread over the Internet. We built up a model which precisely decides if a news article is fake or real using vectorized semantic and syntactical analysis. The codes and results are available at https://github.com/pushkarrrr/Fake-News-Detection/blob/master/fakenews.ipynb.KeywordsNews classificationSyntactical analysisSemantic analysisVectorizationClassification