User Experience (UX) Analysis of Skin Tone Clustering with AI for Clothing Small and Medium Enterprise (SME) System

The difficulty in imagining whether the colour of clothes purchased online will match the skin tone of the buyer or not makes users hesitant and even reluctant to buy fashion clothes from online stores. this is usually influenced by the lighting factor when photographing products, which sometimes produces different results than when they reach the buyer. currently, there are not many websites that provide user experience to overcome this with the help of AI technology, especially skin tone detection from the skin tone of the face directly from the seller's sales platform. Therefore, this research aims to measure the user interface of the clothing colour recommendation system according to skin tone. Since online users can check their skin tone according to the clothes they want to buy, this is expected to improve the user experience of a unique online market share, especially in Kuta Bali, Indonesia. This research uses the software development methodology, SCRUM. This research successfully measures the success factors that affect the user experience in this system and shows the factors in the system that need to be improved before it can be fully used by users. For testing, 3 methods were used, namely User Acceptance Testing (UAT) with an average result of 90.67% (aspects of usability 90%, information reliability 81 % and interaction quality 91 %), for User Experience Questionnaire (UEQ) testing with an average result of 504 (with aspects of attractiveness (1. 699), Clarity (1.632), Efficiency (1.740), Accuracy (1.583), Stimulation (1.588) and Novelty (1.005)), which puts it in the good category, and Regression Testing, which achieved 100% results, showing the reliability and stability of this system. All in all, the development of this website illustrates a user interface designed in accordance with customer expectations.