MUMBAI, India, Sept. 14 -- Intellectual Property India has published a patent application (202641106592 A) filed by Kkr & Ksr Institute Of Technology And Sciences on September 04, 2026, for Agentic Llm System For Automated Project Evaluation.
Inventors include Mr. Prof. Repudi Ramesh; Mrs. Dr. Chittineni Aruna; Kare Achi Rao; Karri Vivek Gopal Reddy; Kambham Sumanth; and Konakanchi Badrinadh.
The application for the patent was published on September 11, 2026, under issue no. 37/2026.
Abstract: ABSTRACT [0012] The rapid growth of academic and industry project submissions has created challenges in ensuring originality, quality, and fair evaluation. Manual project assessment is time-consuming, subjective, and prone to inconsistency, while existing automated evaluation methods often rely on limited rule-based or isolated text-matching techniques that fail to capture deeper semantic meaning and innovation levels. To address these limitations, this invention proposes an agentic large language model system for automated project evaluation. The proposed system extracts key elements from uploaded project documents—including the project title, abstract, problem statement, methodology, and keywords—and converts them into semantic vector embeddings using advanced natural language processing. The extracted content is then compared against summaries and metadata from external repositories such as academic research databases, open-source platforms, and scholarly archives through controlled retrieval mechanisms. A similarity analysis module computes originality scores by measuring semantic alignment between the submitted project and existing works. Based on predefined thresholds, the system categorizes projects as highly similar, moderately similar, or novel. An intelligent agent framework further evaluates project quality using rubric-based scoring criteria, including innovation, technical depth, feasibility, clarity, and impact. Finally, the system generates structured feedback, improvement suggestions, and a standardized evaluation score. Experimental validation demonstrates that the proposed framework enhances fairness, scalability, and accuracy in project evaluation while reducing human bias and evaluation time.
Disclaimer: Curated by HT Syndication.
