MisInfoCheckXAI: Mis-Information Detection Framework using linguistic features and An Explainable AI

The rapid spread of misleading online information requires detection systems that are both accurate and interpretable. This research present MisInfoCheckXAI, an explainable Mis- information detection framework based on linguistic features including Disclosure, Pragmatic, and Integration using Logistic Regression. The model combines TF-IDF representation, SMOTE for data balancing, and SHAP-based explainable AI to provide transparent decision insights. Experimental results achieved an accuracy of 0.8562 with strong macro precision, recall, and F1-score, demonstrating superior classification performance on the machine learning and deep learning approaches, the proposed MisInfoCheckXAI, framework provides a computationally efficient and interpretable solution for trustworthy Mis-information detection using Explainable AI.