The rapid growth of online news platforms and social media has significantly increased the dissemination of misinformation, creating challenges for individuals, organizations, and digital platforms in distinguishing credible information from misleading content. This study proposes a data-driven machine learning framework for identifying misinformation in online news by leveraging natural language processing and supervised machine learning techniques. The proposed framework involves data pre-processing, text normalization, feature extraction, model training, classification, and performance evaluation. Various textual features, including term frequency, word importance, linguistic patterns, and semantic representations, are extracted from online news articles to identify characteristics associated with misinformation. Multiple machine learning algorithms, including Support Vector Machine (SVM), Random Forest (RF), Logistic Regression (LR), and Gradient Boosting, is considered to evaluate their effectiveness in classifying news articles as reliable or misleading. The models are evaluated using standard performance metrics such as accuracy, precision, recall, F1-score, and confusion matrix analysis. In addition, feature importance analysis is incorporated to identify the textual characteristics that contribute significantly to misinformation detection. The proposed data-driven approach aims to provide an automated and scalable mechanism for detecting potentially misleading online news while supporting transparency in the classification process. The findings of this research can contribute to the development of intelligent misinformation-monitoring systems for digital news platforms, social media applications, and information-verification environments.