Development and Evaluation of a Mobile-based Skincare Recommendation System Using Facial Analysis

In this paper, we present a mobile application for personalized skincare product recommendations based on facial skin analysis and real-time environmental conditions.The system integrates four core components: a facial detection module for analyzing skin type and acne severity, a cloud database storing product information and user feedback, a personalized product recommendation module, and an environmental adjustment module that incorporates Global Positioning System (GPS)-based location data and weather application program interface (API) to account for temperature, humidity, and ultraviolet (UV) index.The recommendation algorithm considers not only users' skin type and acne conditions, but also other factors including current location, local climate, previous experience, and crowd-sourced ratings, so the product recommendation can be more personalized and accurate.The user interface was validated through eye-tracking experiments to optimize usability for the public.A 54-user pilot test was conducted to evaluate the user experience.The analysis result showed excellent performance in cognitive load, error handling, and information quality, with users appreciating the contextual relevance of recommendations.This application is particularly suitable for young consumers who shop skincare products independently without the help of professional consultation, and it can serve as a valuable tool in retail environments and online shopping platforms.To coordinate with real-time factors, the system has enhanced the ability to provide personalized, location-based recommendations, filling an important gap in existing skincare recommendation technologies.