NeuroFuzDetect–A Neural and Fuzzy Inference System for Fake News Classification

The widespread dissemination of fake news in the digital era significantly impacts public opinion, economies, and political outcomes. Traditional fake news detection methods often struggle to balance accuracy and interpretability, limiting their effectiveness. To address this challenge, this paper introduces NeuroFuzDetect, an Intelligent Fuzzy-Neural Network that integrates Fuzzy Logic and Long Short-Term Memory (LSTM) to enhance both accuracy and transparency in fake news detection. NeuroFuzDetect leverages advanced computational reasoning to analyze uncertainty and ambiguity in textual data by evaluating critical linguistic features, including credibility, emotional tone, and language patterns, ensuring interpretability by providing justifications for its predictions. The model was evaluated on a widely recognized dataset, demonstrating an accuracy of 87.2%, along with significant improvements in precision, recall, and F1-score. Future research will focus on refining the model by incorporating multimodal analysis, integrating visual and contextual cues, and adapting it for real-time detection on digital platforms while enhancing robustness against adversarial attacks and extending applicability across diverse linguistic and cultural contexts. By achieving a balance between efficiency and transparency, NeuroFuzDetect presents a scalable and reliable solution for combating fake news in real-world applications.