Analysis and Detection of Political Fake News Using Deep Learning with High-Performance Hybrid Model

Fake political news poses a tangible threat to the public consciousness, as it can be manipulated for propaganda purposes and used to obfuscate the truth, thereby influencing societal behavior. The purpose of this study is to improve and develop the performance of deep learning algorithms in detecting fake news. A comparison was made between deep learning algorithms RNN, LSTM, and CNN to analyze data by experimenting with the algorithms individually, then creating a hybrid and sophisticated model to reach a stable and accurate result that is adapted to detect fake political news by scrutinizing over 37 thousand instances of fake political news, sourced from various social media platforms through use of hybrid RNN-LSTM-CNN model. The trained models displayed consistency and reliability, providing an exceptionally robust foundation for future data analysis.