Penerapan Algoritma Hierarchical Clustering untuk Penilaian Pelayanan Akademik Berdasarkan Indeks Kepuasan Mahasiswa

Abstract

In the role of universities to improve the benefit and science to be more developed in the world of technology and development, it is required to produce superior and quality human resources in accordance with the wishes of the community in the field of development. The consumer's benchmark for services is the provider of higher education services as a whole, more than half of the people are now starting to demand quality services and enjoy its convenience. In this study, the questionnaire was used as a benchmark in determining the level of student satisfaction by applying the Hierarchical Clustering Algorithm with 3 types of Hierarchical Clustering algorithm methods, namely single linkage, average linkage and complete linkage. Of the three methods that produce a balanced dendogram, the complete linkage method with objects in each cluster is compared with the single linkage and average linkage methods. For the assessment of academic services based on the student satisfaction index from the overall test data model, the Sum of Squares Error (SSE ) produces the sum of the differences in squares between the predicted data points (ŷi) and the data points. . observed (yi) is 34.67383 with error estimation error accuracy shows good results.