MUMBAI, India, Oct. 5 -- Intellectual Property India has published a patent application (202641114782 A) filed by Vardhaman College Of Engineering on September 25, 2026, for System And Method For Explainable Artificial-Intelligence-Based Decision Support.
Inventors include Mr. J J Dass Kumar; Ms. Hari Durga Srilakshmi; Ms. Gandi Mounika Naidu; Mr. M Bala Subrahmanyam; Dr. Mangalapuri Venkateswarlu; and Dr. Arnab Deb Nath.
The application for the patent was published on October 02, 2026, under issue no. 40/2026.
Abstract: ABSTRACT System and Method for Explainable Artificial-Intelligence-Based Decision Support The present disclosure relates to a system and method for explainable artificial intelligence (AI)-based decision support and adaptive verification of AI-generated decisions. The system gathers heterogeneous input data and associates evidence units with parameters of provenance, reliability and temporal relevance. A dynamic evidence graph links evidence units to derived features, intermediate inference states and an AI-generated decision. The counterfactual perturbation engine changes, hides, deletes or replaces selected evidence, and causes re-execution of an AI inference engine in a comparative fashion. The explanation stability analyser estimates an explanation stability parameter from changes in decision output, evidence contribution and inference states. For an unstable system, an adaptive evidence acquisition controller collects more evidence and a computational-path reconfiguration module selectively reconfigures an inference path or computational resource allocation. The decision is only made when the pre-defined conditions for stability and consistency of evidence are met and it is stored with provenance, perturbation results, verification information, replay data and an integrity value, which improves the reproducibility of the decision, reliability of the verification and computational efficiency.
Disclaimer: Curated by HT Syndication.
