<![CDATA[In daily life, many individuals have difficulty in determining a menu of dishes that match the ingredients available at home. This research aims to develop a web-based cooking recipe recommendation system that is able to provide relevant recipe suggestions based on ingredient input from users. The system is designed by combining the Content-Based Filtering method and the K-Nearest Neighbors (KNN) algorithm to improve the accuracy and relevance of recommendations. Feature representation is performed using the Term Frequency-Inverse Document Frequency (TF-IDF) method to convert the list of materials into a numerical vector form. The measurement of similarity between ingredients and recipes was carried out using the Cosine Similarity approach, while the selection of the best recipe used the top-k technique based on the KNN algorithm. The system was tested against 10 different material input scenarios, and the evaluation results showed that the system had an accuracy rate of 80%, precision of 80%, recall of 100%, and an F1-score of 89%. These findings show that the developed recommendation system is effective in helping users choose recipes that match the available ingredients, as well as potentially reducing food waste. In addition, this system can be a practical solution in supporting the efficiency of household kitchens and encouraging optimal use of materials. Thus, the system has the potential to be further developed and integrated in a digital platform that supports a sustainable lifestyle. ]]>
Copyrights © 2025
