<![CDATA[The rapid growth of social media platforms has significantly increased the spread of misinformation and fake news, which can negatively influence public opinion and social stability. Detecting fake news automatically has therefore become an important research topic in the fields of natural language processing and machine learning. This study proposes a transformer-based fake news detection model using Bidirectional Encoder Representations from Transformers (BERT) to perform semantic analysis on social media news content. The proposed framework consists of several stages including dataset preparation, text preprocessing, contextual feature extraction using BERT embeddings, model training, and performance evaluation. The experiments were conducted using the FakeNewsNet dataset, which contains verified fake and real news articles collected from online sources. The dataset was divided into training and testing subsets, and the model was fine-tuned using a BERT-based classification architecture. The performance of the proposed model was evaluated using standard classification metrics including accuracy, precision, recall, and F1-score. Experimental results show that the proposed model achieved an accuracy of 94.3%, with precision, recall, and F1-score values of 93.7%, 92.9%, and 93.3%, respectively. These results demonstrate that transformer-based language models are highly effective in capturing contextual semantic relationships within textual data. The findings suggest that integrating transformer architectures into misinformation detection systems can significantly improve the accuracy and robustness of fake news detection in social media environments.]]>
Copyrights © 2026
