<![CDATA[The rapidly growing culinary sector requires Micro, Small, and Medium Enterprises (MSMEs) to implement targeted operational strategies. The main problem often faced by the management of the Warung Mie Aceh Seafood MSME is the inability to accurately predict the level of menu popularity based on external factors. This directly impacts the inefficiency of managing wet raw material (seafood) inventory due to its perishable nature. Therefore, this study aims to implement data mining technology using the C4.5 algorithm to predictively classify customers' favorite menus. As a solution to these problems, this study processed 586 historical sales transaction data records from July to September 2025. Unlike previous studies, the novelty of this research lies in integrating external environmental variables, namely weather conditions and day types, along with menu type attributes into a single multidimensional decision tree model evaluated using the 10-fold cross-validation method. The test results show that the C4.5 algorithm classification model has proven to be excellent, achieving an accuracy rate of 96.08% and a class precision value of 100% for the "Very Favorite" and "Not Favorite" classes. Practically, extracting rules from this model provides a specific contribution as a decision support system; for instance, guiding management to reduce cold beverage stocks during heavy rain, or maximize the 'Snacks' menu inventory on weekends, enabling the MSME to avoid material losses while efficiently minimizing the risk of lost potential revenue.]]>
Copyrights © 2026
