Development of the “Scentify” Application as a Web and Mobile-Based Perfume Recommendation System Using Content-Based Filtering Based on User Preferences

This study presents the development of Scentify, a web and mobile-based perfume recommendation system employing the Content-Based Filtering (CBF) algorithm. The research aims to assist users in identifying perfumes aligned with their personal preferences such as fragrance type, concentration, gender, age, and price range. The development methodology includes literature review, user preference data collection via questionnaires, system design, and implementation using Flutter for mobile and Flask for backend API. User preference data were integrated with curated perfume datasets obtained from reliable online sources to form the recommendation base. The results demonstrate that Scentify can produce personalized recommendations with relevant accuracy. It features login, registration, perfume questionnaire, favorites, and admin dashboard modules. The system was validated using black-box testing, proving its reliability and user-friendliness. This work confirms that Content-Based Filtering is effective in building personalized digital recommendation systems and contributes to digital innovation in the perfume and e-commerce industries.