Intelligent Fake News Identification Strategy for Improved Information Authenticity on Online Platforms Using Modern ML and NLP Methods
Muthuswamy Jayanthi, Sajja Suneel, Manish Gupta, Dinesh Kumar Yadav, Mohammed Wael Mohammed, Anas Bilal · 2025
False news spreading on digital channels drastically compromises social stability and faith in information systems. This research paper presents an intelligent methodology for fraudulent news detection utilising state-of-the-art machine learning techniques. Combining deep learning models with natural language processing (NLP) allows the proposed method to exactly identify news articles as true or false. Pre-trained language models-BERT and XLNet among others-are combined in architecture to extract contextual semantic aspects from textual input. Strong feature engineering covering syntactic and linguistic factors is pursued to improve model performance. Training and validation on a huge dataset comprising several sources and languages provides extensive applicability and reliability from the technology. Experimental results reveal that the proposed method achieves greater accuracy than traditional methods even if false positives significantly drop. The paper also explores how two ensemble learning techniques-stack-based and boosting-might assist to raise detection capability even further. Moreover included are real-time deployment scenarios, thereby emphasising the scalability and system interaction with present content management systems of the system. This work supports the ongoing attempts to refute misleading information and thereby helps to preserve the integrity of information dissemination in the digital age by providing a thorough, automated technique for fake news detection.