An LSTM-Based Fake News Detection System Using Word Embeddings-Based Feature Extraction

Abstract

Fake news is manipulated news or misinformation that is spread across the Internet with an intention to impose certain ideas and to damage an agency, organization and person often using dishonest, sensationalist and outright fabricated headlines to increase readership. Due to the propagation of fake news, there is a need for computational methods to detect them. Fake news existed for decades, and in the research community, the detection of fake news has been a desired topic. Around 70% of people are concerned about the propagation of fake news. Given the challenges related to the detection of fake news research problems, the researchers globally are trying to figure out the basic attributes of the problem statement. The objective of this paper is to detect whether the online articles are fake or credible, using various machine learning techniques like GloVe word embeddings and long short-term memory (LSTM) as feature extraction and as a classifier technique to find the best fit for the model.