<![CDATA[Fake news detection has become a major challenge in the rapidly evolving digital era. This study proposes a hybrid approach combining Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to enhance the accuracy of fake news classification. CNN is utilized to extract local features from text, while LSTM captures temporal relationships in sequential data. The dataset used is sourced from Kaggle, consisting of 44,919 fake and real news articles. The classification process includes several stages, such as data preprocessing, tokenization, and transforming text into numerical representations before being processed by the hybrid CNN-LSTM model. Evaluation results indicate that the hybrid CNN-LSTM model achieves an accuracy of 99%, outperforming individual CNN and LSTM models. With high precision and recall rates, this method proves to be effective in classifying fake news, significantly contributing to the development of a more accurate and reliable fake news detection system.]]>
Copyrights © 2025
