Detecting fake news: A hybrid approach combining convolutional neural networks with social learning theory

The proliferation of fake news poses a significant threat to society, undermining trust in information and jeopardizing democratic processes. While the convolutional neural networks (CNNs) have shown promise in detecting fake news by analysing textual features, this study proposes a novel approach that integrates the CNNs with social learning theory. The authors argue that incorporating social signals, such as user credibility, social media engagement and network propagation patterns, can enhance the accuracy of fake news detection. The proposed hybrid model leverages the CNNs to extract textual features and combines them with social signals to capture the influence of social dynamics on the spread and perception of misinformation. We evaluate our model on a real-world dataset of news articles and social media interactions, demonstrating its effectiveness in improving fake news detection accuracy compared to traditional CNN-based approaches. The findings highlight the importance of considering social context in automated fake news detection and offer a promising direction for future researches.