Clusterization of New Student Promotion Strategy Determination at Prisma University Manado Using K-Means Algorithm and Agglomerative Hierarchical Clustering

Abstrak

Prisma Manado University, a newly established higher education institution in 2017, has faced challenges in increasing its number of new students. Due to the increasingly competitive landscape in higher education, determining effective promotional strategies has become essential. In an effort to achieve this goal, this research analyzes the application of Data Mining with a focus on clustering using the K-Means and Hierarchical Clustering algorithms. This study aims to identify groups of prospective students with similar characteristics based on available data, such as educational background, field of interest, and location preferences, so that more targeted promotional strategies can be developed. The K-Means method is used to group prospective new students into clusters based on their similar characteristics. The data used in this study includes demographic data and preferences of prospective students that have been collected by Prisma Manado University. The results of the cluster analysis will provide valuable insights into the preferences and needs of prospective students, which will then help the university in designing more efficient and suitable promotional strategies for each group of prospective students. This helps the university allocate resources more effectively and maximize promotional efforts to achieve new student enrollment targets. By applying the Data Mining approach using the K-Means and Hierarchical Clustering algorithms, this research is expected to provide a deeper understanding of the prospective students of Prisma Manado University and encourage the university to take concrete steps in improving the effectiveness of their promotional strategies. The contribution of this research is to compare two algorithms, namely K-Means Clustering and Agglomerative Hierarchical Clustering.