GlowPick:Cosmetic E-commerce Website with Product RecommendationSystem

The rapid growth of online cosmetic e-commerce platforms has intensified the demand for personalized product recommendations, as generic, popularity-based suggestions often fail to address individual skin characteristics and preferences. While artificial intelligence–based recommendation systems enable scalable personalization, many existing approaches lack transparency, adaptability, and domain-specific safety considerations, particularly in skincare applications. This paper presents GlowPick, an AI-driven cosmetic ecommerce framework designed to deliver personalized and explainable skincare recommendations. The system integrates structured user skin profiling with content-based filtering and rule-driven ingredient compatibility analysis to generate safe and relevant product suggestions. A weighted scoring mechanism is employed to compute product relevance based on skin type, skin concerns, ingredient suitability, and user preferences. The proposed framework is implemented as a full-stack web application and evaluated using a controlled dataset of cosmetic products and simulated user interactions. Experimental results demonstrate that GlowPick achieves higher recommendation accuracy and user engagement compared to traditional recommendation approaches, while maintaining low computational overhead and transparent decision logic. The findings highlight that combining interpretable AI techniques with domain-specific knowledge provides an effective and scalable solution for personalized cosmetic e-commerce platforms.