An Agentic AI Framework for Real-Time Fashion Market Intelligence
Authors:
TEENU JAIN (B. M. Institute of Engineering & Technology)
Shivek Mallick (BM Institute of Engineering and Technology)
Lakshay Rathee (BM Institute of Engineering and Technology)
Dr. Gurminder Kaur (BM Institute of Engineering and Technology)
Dr. Kanika (BM Institute of Engineering and Technology)
Abstract

This paper presents a novel multi-agent artificial intelligence framework designed for real-time fashion market intelligence through automated sentiment analysis and competitive benchmarking. The proposed system employs three specialized autonomous agentsData Collector, Analytics Engine, and Strategic Recommendation Agentoperating in a coordinated architecture to provide continuous market monitoring and actionable business insights. The framework demonstrates significant improvements in processing efficiency (85% accuracy in sentiment classification), response time (30-second cache refresh cycles), and strategic decision-making capability. Experimental results show the system successfully processes multi-platform social media data, competitor pricing intelligence, and brand perception metrics in real-time, achieving 78.4% balanced accuracy in sentiment classification and 40% improvement in adaptive decision-making compared to traditional approaches. The system architecture incorporates advanced natural language processing, hierarchical reinforcement learning, and dynamic visualization components within a Streamlit-based dashboard. Performance evaluation demonstrates robust scalability, autonomous operation, and practical applicability for fashion retail market intelligence.

📄 Download Full Paper (PDF)
Published in: NCAIDT 2025 Proceedings
DOI: 10.63169/NCAIDT2025.p1
Paper ID: NCAIDT2025-0430