MUMBAI, India, Jan. 9 -- Intellectual Property India has published a patent application (202541133415 A) filed by Karpagam Academy Of Higher Education; Karpagam Institute Of Technology; Dr C Sasthikumar; and Jaisurya M, Coimbatore, Tamil Nadu, on Dec. 30, 2025, for 'a secure platform for product authentication and sales approval.'
Inventor(s) include Dr C Sasthikumar; and Jaisurya M.
The application for the patent was published on Jan. 9, under issue no. 02/2026.
According to the abstract released by the Intellectual Property India: "The proposed invention presents an intelligent machine learning-based fraud detection system designed to enhance the accuracy, adaptability, and efficiency of fraud identification in e-commerce transactions. With the increasing volume and complexity of online payments, traditional rule-based systems fail to detect emerging fraudulent tactics that evolve dynamically. To overcome this limitation, the present system integrates supervised and unsupervised learning algorithms, including Decision Trees, Support Vector Machines (SVMs), and Deep Neural Networks (DNNs), for classifying legitimate and suspicious transactions in real time. The architecture consists of multiple interconnected modules, including data ingestion, feature extraction, behavior analysis, anomaly detection, and real-time scoring engines. The data ingestion module collects heterogeneous information from transactional logs, user profiles, device fingerprints, and payment gateways. The feature extraction unit preprocesses and transforms raw data into structured attributes that capture behavioral, spatial, and temporal patterns. The behavioral analysis engine evaluates deviations in transaction frequency, amount, and user behavior to identify potential fraud signals. The fraud detection model combines supervised learning for pattern recognition and unsupervised learning for anomaly identification, enabling detection of both known and previously unseen fraudulent behaviors. A real-time inference engine assigns fraud probability scores to transactions, triggering alerts, holds, or automatic blocking when thresholds are exceeded. The feedback loop continuously updates the model using confirmed fraud outcomes, thereby ensuring adaptive learning and reduced false positives over time. This intelligent detection mechanism is capable of real-time fraud monitoring at scale, while maintaining low latency and high accuracy. It enhances the security of e-commerce platforms by minimizing financial losses and improving user trust. Furthermore, the modular and scalable design enables seamless integration with existing payment systems and supports future advancements in AI-driven cybersecurity and financial fraud prevention."
Disclaimer: Curated by HT Syndication.
