Enhancing Fake News Detection through the Fusion of Classical and Advanced Machine Learning Models

The proliferation of fake news on social networks poses a significant c hallenge to p ublic t rust a nd cybersecurity. This study explores a hybrid approach to enhance fake news detection by integrating classical machine learning models, such as Logistic Regression, AdaBoost, and Support Vector Machines (SVM), with advanced techniques like Long Short-Term Memory (LSTM) and Bidirectional Encoder Representations from Transformers (BERT). Textual content is processed using Term Frequency-Inverse Document Frequency (TF-IDF) and Global Vectors for Word Representation (GloVe) embeddings to analyze syntax and semantics. Experimental results demonstrate that BERT significantly o utperforms o ther m odels, a chieving an accuracy of 97.36%, precision of 95.01%, recall of 99.89%, and an F1-score of 97.39%, positioning it as the most effective solution. These findings u nderscore t he p otential of combining traditional and modern methodologies to strengthen fake news detection systems, fostering a safer digital environment.