A Constructive Content-Based Filtering Recommendation Application: Optimizing Coffee Selection Based on User Preferences

Abstract

Coffee, a typical beverage consumed worldwide, offers various options, from bean origin to brewing methods. However, this abundance of choice often leads consumers to experience decision-making challenges, a phenomenon known as choice overload. Therefore, this study focuses on developing and implementing a recommendation application to help users make informed coffee choices based on their preferences. The study employs a system development life cycle (SDLC) approach and utilizes content-based filtering techniques to achieve the purpose. Data for this study was gathered from three prominent coffee shops: Zus Coffee Shop, Richiamo Coffee Shop and Gigi Coffee Shop, providing a diverse dataset for analysis. The study employs usability testing to evaluate the usefulness, perspicuity, dependability, and attractiveness of the developed recommendation application. Through rigorous testing, we assess user acceptance and overall system performance. The results indicate a significant reduction in choice overload and an enhanced user experience, validating the competence of the recommendation application. The study implies that by using a content-based filtering recommendation system, coffee drinkers can enjoy more personalized suggestions tailored to their taste preferences, such as roast level, milk type, etc. For users who may feel overwhelmed by the variety of options at a coffee shop, this system simplifies the decision process by recommending coffee types that match their stated preferences. By exploring avenues such as collaborative filtering, sentiment analysis, and incorporating additional user feedback, we aim to further enhance the accuracy and personalization of coffee recommendations, ultimately improving the overall coffee selection experience for consumers.