Comparison of single, complete, and average linkage algorithms in clustering instant noodle products based on nutritional content

Abstract

This study aims to perform clustering on instant noodle products based on their nutritional content to assist consumers in making informed choices according to their dietary needs. The method employed is Hierarchical Agglomerative Clustering (HAC), with a comparison of three linkage algorithms: single linkage, complete linkage, and average linkage. The implementation was carried out using the Python programming language on the Google Colab platform. The results indicate that the complete linkage algorithm is the most effective, achieving a silhouette score of 0.4445 and producing two optimal clusters consisting of 98 and 40 products, respectively. These findings suggest that nutritional content–based clustering can serve as a useful reference for providing more informed and responsible instant noodle consumption recommendations.