MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641110607 A) filed by Vallurupalli Nageswara Rao Vignana Jyothi Institute Of Engineering And Technology on September 15, 2026, for An Automated Candidate Selection System Using Nlp-Based Document Understanding, Contextual Embeddings, And Dynamic Matching Algorithms.
Inventor includes Dr. Sagar Yeruva.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: ABSTRACT [0025] An Intelligent Resume Matching System automates the early stages of recruitment by processing PDF resumes and job descriptions through natural language processing and machine learning techniques. The system extracts text content from uploaded PDF documents using specialized parsing tools and generates high-dimensional contextual embeddings for both resumes and job postings via BERT-based sentence transformers. Cosine similarity scores between these embeddings quantify the degree of candidate-job alignment, enabling dynamic threshold-based filtering to shortlist suitable applicants. Shortlisted candidates receive automated congratulatory emails while non-shortlisted candidates receive personalized messages generated by a large language model that identify potential skill or experience gaps and offer constructive guidance. The overall approach reduces manual screening effort, improves matching accuracy beyond simple keyword methods, and maintains positive candidate engagement throughout the process, supporting scalable deployment in modern hiring workflows.
Disclaimer: Curated by HT Syndication.
