Online Fake News Detection

Gautham Murali, Rinsa Rees

Abstract: The proliferation of fake news on digital platforms poses a significant threat to public trust, democratic processes, and societal stability. With the increasing difficulty of distinguishing between authentic and fabricated content, there is a growing need for intelligent systems that can automatically identify deceptive information. This research presents a hybrid approach to fake news detection, combining Natural Language Processing (NLP) techniques with advanced Machine Learning (ML) models to enhance classification accuracy and reliability. The system extracts a range of linguistic, semantic, and metadata features such as sentiment polarity, syntactic patterns, and source credibility from a labeled dataset of real and fake news articles. Multiple classifiers, including Random Forests, Support Vector Machines, and Neural Networks, are trained and optimized using cross - validation techniques. To further improve performance, ensemble learning methods such as boosting and majority voting are employed. The final model is deployed through a web - based application, allowing users to input news content and receive explainable predictions about its authenticity. Experimental results demonstrate high accuracy, precision, and recall, validating the system?s effectiveness in real - world scenarios. This work contributes a scalable and transparent solution to the ongoing challenge of online misinformation.

Keywords: Academic integrity, plagiarism detection, Django, TF - IDF, Cosine similarity, Rabin - Karp algorithm