This research explores the application of a fuzzy logic-based system for detecting fake news on Instagram. The study employs a combination of supervised and unsupervised learning techniques, incorporating features such as title length, description length, likes, comments, source credibility, and author credibility. The supervised approach utilizes logistic regression, support vector machines, and ensemble methods to label posts as “real” or “fake,” while the unsupervised approach leverages fuzzy logic and clustering algorithms to identify latent patterns and anomalies. The fuzzy logic system models the inherent vagueness in fake news classification, enabling a nuanced assessment that traditional binary classifiers might miss. Evaluation of the system demonstrated high accuracy with precision, recall, and F1 scores of 89%, 85%, and 87%, respectively. The results affirm the viability of fuzzy logic in enhancing fake news detection on social media platforms by effectively handling the ambiguity and complexity of online content. Future research could focus on integrating additional features, hybrid models, and real-time detection capabilities to improve performance further.