MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641051176 A) filed by Jennifer D; Ms. R Shakthi Dharshini; Ms. Rakshaka Selvan; Ms. Priyadarsni D; Dr. Valarmathi K; Dr. Umarani Srikanth; Dr. Vijayalakshmi P; Mrs. Preena Jacintha Shalom; Mrs. Jayalakshmi R; and Ms. Soorya S, Chennai, Tamil Nadu, on April 22, for 'blockchain phishing account detector using machine learning.'
Inventor(s) include Jennifer D; Ms. R Shakthi Dharshini; Ms. Rakshaka Selvan; Ms. Priyadarsni D; Dr. Valarmathi K; Dr. Umarani Srikanth; Dr. Vijayalakshmi P; Mrs. Preena Jacintha Shalom; Mrs. Jayalakshmi R; and Ms. Soorya S.
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: "Blockchain technology enables decentralized financial transactions by maintaining transparent and immutable records without relying on centralized intermediaries. Ethereum, a widely used blockchain platform, that supports cryptocurrency transfers and smart contract execution that power decentralized finance applications. However, the open and pseudonymous nature of blockchain networks has also increased the risk of phishing attacks, where malicious actors deceive users into transferring funds to fraudulent wallet addresses. Since blockchain transactions are irreversible, early identification of suspicious wallet behavior is essential to prevent financial loss and strengthen trust within decentralized systems. This work presents a machine learning-based framework for detecting phishing wallet accounts on the Ethereum blockchain through behavioral analysis of transaction data. Blockchain transactions are collected through public APIs and processed to derive wallet-level behavioral features such as transaction frequency, interaction diversity, entropy of outgoing transfers, and fund movement patterns. These features are used to train a supervised learning model that classifies wallet addresses as legitimate or malicious. In addition, the system assigns a probabilistic risk score to each wallet, enabling continuous monitoring and flexible risk assessment. Experimental evaluation shows that the proposed framework effectively distinguishes phishing wallets from legitimate accounts. The results demonstrate that combining behavioral blockchain analytics with machine learning provides a scalable approach for improving fraud detection and enhancing security in decentralized financial environments."
Disclaimer: Curated by HT Syndication.
