Design and Optimization of a Single-Tenant E-Commerce Platform with AI-based Product Recommendations

Abstract

This thesis presents the design and optimization of a single-tenant e-commerce platform with a focus on practical performance and usability. The recommendation system is based on user behavior, particularly tracking clicks and interaction patterns, which provides a straightforward yet effective personalization mechanism. Instead of relying on complex machine learning models, the work emphasizes a lightweight and efficient approach suitable for real-world deployment. The system architecture is clearly structured, ensuring maintainability and scalability within a single-tenant environment. Optimization techniques applied to backend processes and database queries improve overall system responsiveness. The implementation demonstrates a solid understanding of how user activity can be leveraged to enhance the shopping experience. Overall, the thesis delivers a practical and efficient solution, balancing simplicity with functional value in e-commerce systems.