MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641044289 A) filed by Sendhan Thirumanam Umapathy; V Dharshan; V Karthikeyan; SS Rachna; and Dr. R. Dayana, Chennai, Tamil Nadu, on April 7, for 'a framework for detecting zero-day vulnerabilities in e-commerce supply chain.'
Inventor(s) include Sendhan Thirumanam Umapathy; V Dharshan; V Karthikeyan; SS Rachna; and Dr. R. Dayana.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "This patent discloses a system for detecting intrusion into an e-commerce (retail) supply chain with a focus on detecting zero-day vulnerabilities, using-a hybrid of tniditionallclassical machine teaming algorithms (i.e., Random Forest and XGBoost) and newer/deep learning methodologies, specifically Graph Attention Networks (GATs) to identifY anomalous behavior ofinterrelated nodes in the supply chain. Machine learning algorithms are used to understand the interactions between data sources (thirdparty API gateways, logistics databases and payment processors) and to uncover the complex relationships and structures used by signature-based IDS in discovering malicious deeds that cannot be discovered using traditional signatures (zero-day exploits). A combination of the majority vote of all classifiers, using all of the infonnation available (including all known security threats and zero-day exploits), results in a complete security console that provides real-time visibility into the supply chain."
Disclaimer: Curated by HT Syndication.
