Fake News Classification Using Machine Learning and Deep Learning: A Comparative Approach

The exponential expansion of online networking platforms has intensified the circulation of misleading content, impacting community confidence and public understanding. Conventional identification methods frequently struggle to recognize linguistic and situational characteristics within extensive digital material volumes. This research presents a False Information Detection framework that combines Machine Learning with Deep Learning methodologies. Logistic Regression and Support Vector Classifier algorithms served as foundational approaches, while a combined CNN-LSTM architecture was constructed to improve effectiveness through merging pattern recognition with sequence modeling capabilities. Our investigation revealed that the CNN-LSTM system delivered optimal results, reaching 87.78% classification accuracy. Findings demonstrate that this developed approach offers a dependable and expandable answer for identifying deceptive content and minimizing false narratives across internet-based communication channels.