MUMBAI, India, July 24 -- Intellectual Property India has published a patent application (202641085253 A) filed by Dr. D. Maria Sahaya Diran; G Niveditha; Dr. Subba Rayudu Thunga; Yedla Chandini; Mutyala Satish; Pradeepa B; Dr. S. Velayutham; Gomathi M; R. Sathishkumar; Dr. S. Jabeen Begum; Dr D J Samatha Naidu; and Ram Nivas Duraisamy on July 12, 2026, for Cloud-Iot Integrated Machine Learning Framework For Real-Time Credit Card Fraud Detection And Secure E-Commerce Transaction Authentication.
Inventors include Dr. D. Maria Sahaya Diran; G Niveditha; Dr. Subba Rayudu Thunga; Yedla Chandini; Mutyala Satish; Pradeepa B; Dr. S. Velayutham; Gomathi M; R. Sathishkumar; Dr. S. Jabeen Begum; Dr D J Samatha Naidu; and Ram Nivas Duraisamy.
The application for the patent was published on July 17, 2026, under issue no. 29/2026.
Abstract: Credit card payments have become one of the most common ways to make financial transactions due to the rapid increase in digital banking, internet shopping, mobile payments and contactless transactions. The growth, however, comes at a cost as financial fraud is becoming more advanced and thus transaction security has become one of the primary concerns for banks, merchants and consumers. Current fraud detection services largely rely on pre-determined rules, static security policies or few transaction attributes to detect anomalies. While these methods can catch specific frauds, they often do not recognize novel orders that have recently arrived, leading to detection delays as well as the needless blocking of legitimate transactions—both creating customer annoyance. Current systems also tend to work in silos, not leveraging the massive amount of information that is generated via connected payment devices, cloud platforms and customer behaviour over time, handcuffing their ability to perform intelligent and adaptive fraud prevention. Accordingly, the present invention is concerned with a Cloud-IoT Integrated Machine Learning Framework for Real-Time Credit Card Fraud Detection and Secure E-Commerce Transaction Authentication which provides an all-in-one framework that integrates cloud computing, IoT-enabled payment infrastructure, behavioral feature engineering, machine learning and adaptive authentication. The system always collects data about transactions from banking systems, payment gateways, merchant platforms, customer devices, and terminals for contactless payments. The data collected is then distilled into useful behavioral parameters such as the Transaction Trust Score (TTS), User Behavioural Consistency Index (UBCI), Device Reliability Factor (DRF) Location Confidence Score (LCS) and Dynamic Fraud Risk Index (DFRI). Through a machine learning prediction engine, these indicators are then processed by the banks to determine fraud probability, detect unusual transaction behavior and aid in making intelligent authentication decisions. The framework also emphasizes the utilization of features such as monitor them in real time, receive fraud alerts, offer adaptive authentication solutions and continuously adapt to new patterns with machine learning algorithms that yields better detection performance when more information about the transactions becomes available. With this invention, businesses can expect improvements in transaction security, reduction in erroneous decline of legitimate transactions, improved fraud detection performance, and building stronger customer trust; as well as a scalable digital payment ecosystem to be merged into future eCommerce platforms. FIG.1
Disclaimer: Curated by HT Syndication.
