MUMBAI, India, June 22 -- Intellectual Property India has published a patent application (202641048355 A) filed by Saveetha Engineering College on April 16, 2026, for Hybrid Generative Ai Framework For Fake News Detection In Regional Media.
Inventor includes N. Madhumitha.
The application for the patent was published on June 12, 2026, under issue no. 24/2026.
Abstract: The rapid expansion of digital media platforms has significantly increased the spread of misinformation, particularly in regional languages where automated detection systems remain underdeveloped. This paper proposes a Hybrid Generative Artificial Intelligence Framework for Fake News Detection in Regional Media, designed to address linguistic diversity, contextual ambiguity, and the lack of explainability in existing systems. The proposed system integrates generative AI models (such as Large Language Models) with discriminative machine learning classifiers to create a robust and interpretable detection mechanism. The framework processes multilingual and code-mixed inputs commonly found in regional media, including Tamil, Hindi, and other low-resource languages. It employs advanced Natural Language Processing techniques such as contextual embeddings, semantic analysis, and transformer-based architectures to understand nuanced meanings and detect subtle misinformation patterns. A key feature of the system is the incorporation of Retrieval-Augmented Generation (RAG), which enables real-time fact verification by retrieving trusted information from reliable sources and comparing it with the input content. Additionally, the framework includes an Explainable AI (XAI) module that generates human-readable justifications for its predictions, thereby improving user trust and transparency. The system is further designed to support multimodal data, including textual content, images, and metadata, enhancing its capability to detect fake news in diverse formats such as social media posts and news articles. Its adaptive learning mechanism ensures continuous improvement by incorporating new data and evolving misinformation trends . Overall, the proposed framework provides a scalable, accurate, and user-centric solution for combating fake news in regional media ecosystems, making it highly suitable for deployment in real-world applications such as social media monitoring, news verification platforms, and mobile applications.
Disclaimer: Curated by HT Syndication.
