AI-Enabled Conversational Commerce: A Smart Chatbot Framework for E-Commerce Websites

Artificial intelligence is increasingly being used to make e-commerce interactions faster, more personalized, and less dependent on conventional customer-support channels. This paper proposes a smart conversational framework for an e-commerce website in which an AI chatbot acts as an interactive layer between the customer, product catalogue, order services, recommendation engine, and feedback analytics. Unlike a basic FAQ chatbot, the proposed framework is designed to understand customer intent, retrieve product information, support product discovery, assist with cart and order-related queries, collect feedback, and use interaction signals to improve future responses. The design combines natural language processing, intent classification, entity extraction, catalogue search, recommendation logic, and sentiment analysis in a modular architecture. A conversation-to-commerce feedback loop is introduced so that chatbot interactions can generate useful non-sensitive service signals for improving product recommendations and support quality. The paper presents the system architecture, functional modules, processing workflow, database requirements, evaluation metrics, security considerations, and implementation roadmap. The work is positioned as a practical and extensible framework for small and medium e-commerce platforms. It does not claim experimental performance without a real deployment; instead, it defines measurable indicators such as intent accuracy, response relevance, task completion rate, response time, customer satisfaction, and recommendation acceptance. The proposed approach provides a foundation for developing an e-commerce chatbot that supports both customer experience and data-informed business improvement