Fake News Detection using Hybrid BiLSTM-TCN Model with Attention Mechanism

In the past decade, social media has seen a tremendous surge in popularity, but it has also come with its fair share of problems. One of the most significant concerns is the proliferation of disinformation, or false and misleading information spread by individuals and organizations for the purpose of promoting their own ideologies and products. To combat this rise of disinformation, several text detection methods based on artificial intelligence and machine learning have been developed. This study presents a novel solution using a combination of Bi-LSTM, TCN, and attention mechanism. The proposed model leverages Glove Embedding for word representation, Bi-LSTM for capturing sequential text dependencies, attention mechanism for identifying important text elements, and TCN for deep feature extraction. The efficacy of the proposed model was assessed through empirical trials utilizing several models on the publicly available ISOT dataset. The results of these trials were highly impressive, demonstrating success of this research endeavor.