MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202611088600 A) filed by Manipal University Jaipur on July 21, 2026, for A Multimodal Spatio-Temporal Deep Learning System For Predictive Financial Risk Detection.
Inventor includes Dr Sonal Sidana.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: The present invention relates to a system and method in the field of artificial intelligence and financial data analytics. The system employs a hybrid deep-learning architecture comprising a Convolutional Neural Network (CNN) and a Long Short-Term Memory (LSTM) network for processing multimodal financial data and identifying temporal and spatial patterns associated with financial behaviour. The system receives and preprocesses heterogeneous financial data comprising transaction, behavioural, temporal, spatial, and other relevant financial information, extracts spatial and feature-level representations using the CNN, and analyses sequential dependencies and time-dependent behavioural patterns using the LSTM network. The processed representations are combined to generate predictive outputs indicative of financial anomalies, fraudulent activities, behavioural deviations, and financial risk. The invention enables pre-event or predictive analysis of financial behaviour, thereby improving upon conventional reactive fraud and risk detection techniques. The disclosed architecture provides improved accuracy, efficiency, and predictive capability in identifying financial risks and anomalous financial behaviour through integrated deep-learning-based analysis of multimodal and sequential financial data.
Disclaimer: Curated by HT Syndication.
