Skincare Recommender System Using Content-Based Filtering Method

Skincare has recently emerged as one of the fastest-growing product categories, driven by increasing consumer awareness and demand for personalized beauty solutions. In the past, individuals often selected skincare products based on best-selling items, word-of-mouth recommendations, or suggestions provided by retail stores. To assist potential consumers in choosing the most suitable product, an effective recommendation system is essential. This study proposes a system that provides skincare product recommendations based on products that have been used by consumers. The method used is content-based filtering, which compares product characteristics to produce more relevant recommendations. The results show an F1measure of 0.87, and an accuracy of 78.78%. In addition, the system reach precision@50 of 0.98, recall@50 of 0.7656, and mean absolute error (MAE) of 0.21. These results reflect the system's strong performance in aligning recommendations with user preferences. The recommendation system can improve the consumer experience by simplifying the decision-making process and offering suggestions tailored to individual skincare needs.