Paper Title
Extraction of Significant Information from Documents using LLM

Abstract
Information extraction’s traditional methods are time-consuming and inflexible. Parsers automate the process but struggle with document format variations. NLP with dictionaries improves speed but lacks adaptability to different phrasings. Our approach utilizes LLM-generated word Embeddings for classification documents and information extraction. This allows the model to handle various sentence structures and improve accuracy by considering multiple word usages. The extracted information is then verified to gauge model effectiveness. This method offers a more robust and adaptable solution for information extraction from unstructured documents. Keyword - Resume Analyzer, Applicant Tracking System (ATS) Lite, Resume Skill Extractor, Candidate Information Extraction Tool