MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202611090808 A) filed by Noida Institute Of Engineering And Technology Niet on July 26, 2026, for System And Method For Predictive Assessment Of Employee Innovation And Adaptive Manager Intervention.
Inventors include Pooja; Taanya Seth; and Annu Sharma.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: A hardware-anchored, privacy preserving computer implemented system (100) and method are disclosed for predictive assessment of employee innovative behavior and adaptive orchestration of manager interventions. The system (100) comprises endpoint devices with trusted execution environments hosting a multi-source data ingestion subsystem (110) and a privacy preserving federated layer (190) employing additive homomorphic masking and differential privacy; a feature engineering module (120); a temporal sequence encoder (130); a hybrid predictive engine (140) with gradient boosting (142), graph neural network (144) and attention fusion (146) branches executing on a graphics processing unit accelerated inference server; an innovation likelihood scorer (150); an intervention recommendation engine (160) employing a doubly robust causal inference sub-module (162); a manager dashboard (170); a conservative Q-learning feedback loop (180); and a data repository (195). The system produces a calibrated forward innovation probability and measurable uplift in innovation rate.
Disclaimer: Curated by HT Syndication.
