A Real Time AI System For Stock Analysis And Investment Support
B Niharika,
B Reethu Sahithi,
G Sreeshanth,
S Yuvateja,
S Mahesh
The Indian stock market, comprising the National Stock Exchange (NSE) and the Bombay Stock
Exchange (BSE), has witnessed unprecedented retail investor participation, with over 192 million
registered demat accounts as of 2025. Despite this growth, the vast majority of retail participants lack
access to structured, institutional-grade equity research. Investment decisions remain heavily influenced
by unverified market tips, social media narratives, and speculation rather than systematic analysis.
This paper proposes a novel Multi-Agent Artificial Intelligence (AI) framework that democratizes
institutional-quality financial research for Indian retail investors. The system deploys five specialized
LLM-powered agents—covering fundamental analysis, technical indicators, macroeconomic context,
news sentiment, and risk assessment—whose outputs are aggregated by a synthesis agent to produce
BUY, SELL, or HOLD recommendations with quantified confidence scores, and are compiled into
professional PDF research reports delivered via a secure web interface. Experimental evaluation on
Nifty 50 and BSE Sensex stocks achieves a recommendation accuracy of 76.4%, outperforming singleagent baselines by 18.3 percentage points.
