<![CDATA[The rapid growth of the food and beverage industry in Indonesia and the increasing number of available menu options have given rise to the paradox of choice phenomenon, which causes consumers difficulty in making effective decisions. This condition drives the need for a recommendation system capable of helping users find menu options that match their nutritional needs and preferences. This study develops a website-based food menu recommendation system using the K-Means clustering algorithm to group menus based on nutritional content, including calories, protein, fat, and carbohydrates. The system development method uses the Waterfall approach with stages of requirements analysis, system design, implementation, testing, and maintenance. The system is built using the PHP Laravel framework and MySQL database. Black box testing results show that all system functions operate according to specifications with a 100% success rate. White box testing on the K-Means algorithm produces a cyclomatic complexity value of 6 for the clustering process and 2 for the recommendation search process, indicating a logical complexity that remains within reasonable bounds. The system successfully grouped 15 local Indonesian food menus into 3 clusters based on nutritional value similarity. For future research, it is recommended to add a geofencing feature so recommendations also account for the user's proximity to food providers, to implement server-side data caching to optimize loading speed as menu data grows, and to conduct ongoing research on user behavior to adjust the weighting of each menu characteristic in order to maintain the relevance of the recommendations produced.]]>
Copyrights © 2026
