Two-Way Truth Seeker: A Hybrid Method Using LSTM and BiLSTM to Recognize and Classify Fake News

Abstract

In the era of digitization, news spreads, which has turned out to be a big challenge because people seem to be being systematically misinformed. This lowers the credibility of information given on the web and brings threats against social stability BiLSTM and LSTM fusion for fake news classification: Two-way truth seeker. The study of this research focuses on developing an improved advanced model of deep learning that has been trained on a large dataset of labeled news which increases its ability to efficiently detect fake news. The system utilizes the concept of a BiLSTM and LSTM network fusion approach which enables the process to efficiently capture long-range dependency from forward and backward directions to improve the detection of false information. As a result, the presented combination provides improved accuracy and robustness in detecting fake news. Extensive experiments and evaluations were performed using a diverse set of news articles. This project will not only help in the domain of fake news detection but also prove how well-advanced deep learning works on solving the real-world problem.

Keywords

• fake news detection
• deep learning
• lstm
• bilstm
• ensemble model
• natural language processing
• sentimental analysis