Segmentasi dan Persepsi Mahasiswa Program Studi Manajemen Universitas Bunda Mulia terhadap Proses Pembelajaran dengan Metode Cluster Analysis\ud

Although universities are increasingly competing for accepting huge number of students, little has been written about a popular alternative to the lecture-oriented approach to teaching the students at the universities. This paper sees how educational philosophies that underlie lecture and case methods of teaching are related to how faculty perform their teaching to their students. Differences between the underlying world of views of lecture and case methods of teaching similarly lead to differences in many other aspects of the teaching and learning process. The findings of this paper are that each educational philosophy favors a certain instructional methodology, which in turn determine not only the way the instruction is performed but also how faculty perform their teaching, which will be accepted as the most favorable teaching method by the students. These studies investigate the values of students’ perception in accepting the teaching methods as well as the most favorable teaching methods for both qualitative subject and quantitative subject lectures. The way segmentation of the students analyses are performed, reflecting the perception of similarity and difference of the segmentation groups.\ud A new development in segmentation analysis is the cluster analysis method. This method defines segment on the basis of their similarity and difference of response. The K-means cluster analysis algorithm has been a widely applied clustering technique, especially in the area of marketing research. In spite of its popularity and ability to deal with large volumes of data quickly and efficiently, K-means cluster analysis to yield improved performance in terms of solution quality and robustness. Among the algorithms, K-means is one of the most popularly used in the area of marketing research to partition objects into different clusters, due to its ability to deal with large-sized problems that often characterize many marketing research studies. Although K-means is iterative in nature and requires a large amount of computation time to converge, it is more efficient than most other methods in forming clusters, and therefore makes the computational problem more manageable.\u