This paper presents the design and implementation of an Intelligent Shopping Recommendation System that addresses the increasing need for personalized product suggestions in modern e-commerce environments. The system combines a React-based responsive front-end with real-time multi-platform data crawling and advanced AI-driven recommendation generation. By integrating large language models such as Qwen and DeepSeek, the system supports both keyword-based search and conversational interaction, allowing users to refine their shopping needs through natural dialogue. Functional and performance tests confirm that the system delivers real-time, contextually relevant recommendations and stable search results across mainstream desktop environments and browsers. The architecture is modular and lightweight, leveraging serverless deployment and external AI services for scalability and maintainability. While the current implementation focuses on selected e-commerce platforms, future work will expand data sources and enhance personalization through user behavior modeling and continuous learning. The study demonstrates the feasibility and practical value of combining modern web technologies with AI capabilities to improve shopping efficiency and user satisfaction, providing a foundation for future intelligent retail solutions that adapt to evolving technologies and consumer expectations.
Bohan Wu (2025) studied this question.