ARCHIVES

With the rapid spread of misinformation through online platforms, detecting fake news has become a critical challenge for both users and automated systems. This paper presents an approach for classifying news articles as fake or real by leveraging a combination of classical machine learning and deep learning techniques. A dataset comprising thousands of labeled news articles was preprocessed through standard text cleaning methods, including lowercasing, punctuation removal, stopword elimination, and lemmatization. Feature extraction was performed using the Term Frequency-Inverse Document Frequency (TF-IDF) method, limited to 5000 features to ensure computational efficiency. Multiple machine learning models—Logistic Regression, Naïve Bayes, Support Vector Machine, and Random Forest—were trained and evaluated using accuracy, confusion matrices, and classification reports. In parallel, deep learning architectures, including Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), and Long Short-Term Memory (LSTM) networks, were implemented to capture semantic patterns within the text. Experimental results demonstrate that the CNN model achieved superior performance, outperforming other models in both accuracy and robustness. The findings suggest that deep learning models, particularly CNNs, are highly effective for fake news classification tasks when combined with appropriate feature extraction methods. The paper concludes by discussing the implications of these results and identifying directions for future research, including the application of these models to real-time news streams and their broader impact on information credibility.