The explosive growth of digital platforms has intensified the circulation of false information, undermining public confidence and societal harmony. Conventional machine learning and deep learning methods often lack transparency and struggle with vague or noisy text. This study introduces a Fuzzy Neural Network that merges neural adaptability with fuzzy logic’s ability to manage uncertainty. Linguistic, semantic, and credibility features are modeled through fuzzy membership functions and refined via neural learning. Evaluation on standard datasets shows improved accuracy and clear, interpretable decisions, supporting dependable real-world misinformation detection.