LLM based Information Extraction and Retrieva...

LLM based Information Extraction and Retrieval for Finance News Classification and Analysis using NLP

Author Name : Mr. Aftab Malik, Dr. Manoj Sabnis

DOI: https://doi.org/10.56025/IJARESM.2025.1302252351

ABSTRACT This research presents a financial information retrieval system designed to assist Equity Research Analysts (RAs) by leveraging Large Language Models (LLMs) for efficient query processing. The system addresses key challenges, including time-consuming manual analysis, fragmented data across multiple financial articles, and token limitations in conversational AI systems like ChatGPT. The purpose of this study is to develop a solution that integrates dynamic query handling, multi-document aggregation, and user-friendly interaction through a web-based interface. The methodology employs advanced Natural Language Processing (NLP) techniques, including domain-specific LLM adaptation and vector-based retrieval using FAISS, to enable context-aware, accurate responses to complex financial queries.