MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641109541 A) filed by Kuldeep Chowdary Raavi; Meher Deepika Uppaluri; Aravind Kumar Karpoorapu; and Srilekha Vuyyuru on September 11, 2026, for A Hybrid Neuro-Symbolic System For Explainable Decision Validation In Enterprise Artificial Intelligence.
Inventors include Kuldeep Chowdary Raavi; Meher Deepika Uppaluri; Aravind Kumar Karpoorapu; and Srilekha Vuyyuru.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: The present invention discloses the development of a hybrid neuro-symbolic framework for explainable decision validation in enterprise AI settings. In the process, the method merges neural learning methods with symbolic reasoning, business rules, domain knowledge, and logic to enhance the explainability, reliability, and validation of AI decisions. The method accepts enterprise structured and unstructured data as input, performs the processing of the data using neural inference engines, and creates predictive decisions with confidence measures attached to them. The symbolic reasoning module then validates the decisions generated by the system according to rules, policies, regulations, and other constraints set in place. A separate explainability module analyzes the data and reasoning process to find important data features, reasoning, rules used, and potential errors, thereby generating human-understandable explanations. FIG.1
Disclaimer: Curated by HT Syndication.
