MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641052131 A) filed by Nandha Engineering College on April 24, 2026, for An Intelligent Speech Recognition System For Sign Language Visualization.
Inventors include J Jenshya; A Jamuna; G Nandhini; R Sujaypraveen; and M Thamizharasan.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: ABSTRACTOFTHEINVENTION The rapid growth of digital payment systems and online financial services has significantly increased the volume and velocity of credit card transactions worldwide. While these advancements have improved convenience and accessibility, they have also led to a sharp rise in sophisticated credit card fraud, posing substantial financial losses and security threats to consumers and financial institutions. Conventional credit card fraud detection systems on centralized machine learning models that require aggregating sensitive transaction data from multiple sources. This project proposes a Privacy-Preserving Split Learning Framework for Real-Time Credit Card Fraud Detection Using Hybrid Al Models. The proposed framework leverages split learning, a distributed machine learning in which the neural network is divided between the client and server.By preventing raw transaction data from leaving the client environment, the framework ensures strong data privacy and significantly reduces the risk of data leakage while maintaining collaborative model training. In summary, this project presents a novel and practical approach to credit card fraud detection by combining privacy-preserving split learning with hybrid Al models. The results highlight the potential of split learning as a viable alternative to traditional centralized and federated learning approaches in privacy critical applications, paving the way for secure, scalable, and intelligent fraud detection infrastructures.
Disclaimer: Curated by HT Syndication.
