Detecting the Deception : An Intelligent Machine Fake News Detection

The current landscape of digital and social media has made any news universally available. Yet, today's media is marred by dissemination of fabricated news. Known as fake news, it erodes the trust of society, leads to social disorder, and hinders social order. This paper presents a machine learning methodology to detect fake news. Natural language processing (NLP) is a branch of computer science that aids in detection of fictitious news. In this case, NLP is used to pre-process news articles. This includes, but is not limited to, text completing, tokenization, the elimination of stop words, and feature extraction by means of term frequency and inverse document frequency (TF-IDF). News articles are then classified as real or fake by various machine learning classification algorithms. These include, but are not limited to, logistic regression, Naïve Bayes, random forest, and support vector machines. A variety of metrics are used to determine the efficacy of each algorithm. These include accuracy, precision, recall, and the F1-score. The machine learning and NLP methods presented in this paper prove that they have the capability to identify misleading news, and this work provides a starting point for machine learning and NLP methods in fictitious news detection. It also describes the hurdles in reliable fictitious news detection and identifies areas for future research in improving reliable and effective fictitious news detection frameworks.