Unmasking Misinformation: A Transparent Machine Learning Approach to Fake News Detection

The growing influence of online news platforms has amplified the spread of misinformation, making automated fake news detection more critical than ever. This research presents a machine learning-based framework that not only classifies news articles as real or fake with high accuracy but also provides interpretable insights into the decision-making process. We preprocess raw text data using TF-IDF vectorization and train multiple models, including Logistic Regression, Naive Bayes, and Decision Tree classifiers. To ensure transparency, we integrate explainable AI techniques such as SHAP and LIME, allowing us to understand which words influence model predictions most. Experimental results demonstrate strong performance, with Logistic Regression achieving the best balance of accuracy and interpretability. This work emphasizes the importance of explainability in AIdriven content verification, especially in highstakes information environments.