<![CDATA[Increasing ultraviolet radiation exposure and the low adoption of sunscreen among men, compounded by difficulties in selecting products suitable for different skin types, highlight the need for a more objective and personalized recommendation system. This study aims to develop a web-based Artificial Intelligence Decision Support System (AI-DSS) using the Preference Selection Index (PSI) method to recommend sunscreen products according to users' skin types. A Research and Development approach with the Extreme Programming (XP) model was employed, involving a survey of 249 male students, evaluation of 10 sunscreen alternatives using five criteria, and system validation through Root Mean Square Error (RMSE). The results indicate that Active Ingredient Count (0.598) and Product Label Claims (0.536) are the most influential criteria, while Amaterasun UV Sunscreen Serum achieved the highest preference score (0.95). The system achieved an RMSE of 0.0089, demonstrating excellent computational accuracy. This study contributes an objective, accurate, and data-driven AI-based recommendation system to support personalized sunscreen selection. Abstrak Paparan radiasi ultraviolet dan rendahnya penggunaan sunscreen pada laki-laki, yang diperparah oleh kesulitan memilih produk sesuai tipe kulit, menunjukkan perlunya sistem rekomendasi yang lebih objektif dan personal. Penelitian ini bertujuan mengembangkan Artificial Intelligence Decision Support System (AI-DSS) berbasis web menggunakan metode Preference Selection Index (PSI) untuk merekomendasikan produk sunscreen berdasarkan karakteristik tipe kulit pengguna. Penelitian menggunakan pendekatan Research and Development dengan model Extreme Programming (XP), melibatkan survei terhadap 249 mahasiswa, evaluasi 10 alternatif sunscreen berdasarkan lima kriteria, serta validasi sistem menggunakan Root Mean Square Error (RMSE). Hasil penelitian menunjukkan bahwa Active Ingredient Count (0.598) dan Product Label Claims (0.536) merupakan kriteria paling berpengaruh, sedangkan Amaterasun UV Sunscreen Serum memperoleh nilai preferensi tertinggi (0.95). Nilai RMSE sebesar 0.0089 menunjukkan akurasi sistem yang sangat tinggi. Penelitian ini berkontribusi menyediakan sistem rekomendasi sunscreen yang objektif, akurat, dan mendukung pengambilan keputusan berbasis data.]]>
