Intellectual Property India Publishes Patent Application for An Adaptive Neuro-Behavioral Analytics Platform Using Deep Learning And Synthetic User Simulation For Insider Threat Detection Across Cloud, Iot, And Privileged Access Networks

MUMBAI, India, Aug. 12 -- Intellectual Property India has published a patent application (202641092419 A) filed by Sr University on July 30, 2026, for An Adaptive Neuro-Behavioral Analytics Platform Using Deep Learning And Synthetic User Simulation For Insider Threat Detection Across Cloud, Iot, And Privileged Access Networks.

Inventors include Deepthi Bolukonda; Dr. Rupesh Kumar Mishra; and Dr. Indrajeet Gupta.

The application for the patent was published on August 07, 2026, under issue no. 32/2026.

Abstract: Disclosed is an adaptive neuro-behavioral analytics platform for insider threat detection across cloud, IoT, and privileged access networks, employing deep learning and synthetic user simulation. The platform comprises a multi-source data ingestion module that normalises heterogeneous telemetry from cloud audit logs, IoT device gateways, and privileged access management systems, and a neuro-behavioral feature extraction layer that converts such telemetry into dense behavioral embeddings for each monitored identity. A synthetic user simulation engine, employing generative adversarial and agent-based techniques, generates statistically realistic synthetic behavioral data representing both benign and malicious personas, augmenting scarce authentic training data and mitigating class imbalance. A hybrid deep learning inference layer, combining recurrent and attention-based neural architectures, establishes continuously updated behavioral baselines and computes composite risk scores by comparing real-time embeddings against such baselines, weighted by resource sensitivity and privilege context. Dedicated modules address command-level analysis of privileged sessions and device-behavior modelling for IoT fleets. An adaptive retraining subsystem incorporates analyst feedback to refine detection accuracy over time, while an explainable alerting and orchestration layer presents prioritised alerts and interfaces with existing SIEM, SOAR, and identity management infrastructure, thereby improving detection accuracy and reducing false positives relative to prior art systems.

Disclaimer: Curated by HT Syndication.