MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202621056227 A) filed by Dr. Sushil Kumar Gupta; Dr. Pradip Padhye; Dr. Shiba Prasad Mohanty; Dr. Nilesh Kate; Dr. Jairaj Sasane; Dr. Harshal Raje; and Dr. Neha Parashar on May 03, 2026, for A System And Method For Machine Learning-Based Loan Default Prediction Using Adaptive Risk Modeling And Explainable Ai.
Inventors include Dr. Sushil Kumar Gupta; Dr. Pradip Padhye; Dr. Shiba Prasad Mohanty; Dr. Nilesh Kate; Dr. Jairaj Sasane; Dr. Harshal Raje; and Dr. Neha Parashar.
The application for the patent was published on July 03, 2026, under issue no. 27/2026.
Abstract: The present invention relates to a system and method for predicting loan default risk using a machine learning-based framework integrating adaptive risk modeling and explainable artificial intelligence. The system collects and processes heterogeneous data from multiple sources, including financial, transactional, and behavioral inputs, and transforms the data through preprocessing and feature engineering techniques to generate meaningful predictive attributes. A hybrid machine learning model comprising ensemble and deep learning algorithms is employed to accurately estimate the probability of borrower default. The invention further incorporates an adaptive learning mechanism for continuous model updating using real-time data, ensuring sustained performance under dynamic conditions. An explainability module provides interpretable insights into prediction outcomes, enhancing transparency and regulatory compliance. The system outputs a risk score along with classification and confidence metrics, thereby enabling efficient, data-driven decision-making for financial institutions. Accompanied Drawing [FIGS. 1-2]
Disclaimer: Curated by HT Syndication.
