Segmentation of Students Using K-Means Clustering Case Study of Sekolah Dasar Tahfizh Quran (SDTQ) Citamulia

Abstract

This study aims to determine consumer segmentation using K-Means Clustering in the case of SDTQ Cita Mulia elementary school students. The results of the study can be useful for preparing marketing strategies and sharpening the school's competitive advantages. The process consists of grouping consumer data based on several variables, which are student’s gender, distance from home to school, age of student parents, parents' occupation and income, parents' education and the curriculum used by the previous school. After grouping the data, the initialization stage is carried out by giving an initial number. Then the clustering process is carried out. The resulting cluster will be analyzed to see the current Student profile. By studying the clusters formed, it helps us to determine school development policies, especially how marketing strategies can attract consumers in the middle cluster, namely those who have not joined us much but have the potential to be interested in our products. The result of the process there are four clusters with each cluster has specific persona name as follows: The first Cluster is the reachable distance with 5-6 km distance from the school, the second cluster in The long distance with  6 up to 28 Km distance from the school, the third cluster is the  neighbor with just 3-4 km from school, and the forth cluster is the have this cluster persona has the highest parents income with the distance from school is 5-6 Km. The largest distribution of members is cluster 3 followed by cluster 4. Meanwhile, cluster 1 is not too big yet but still looks potential, and the lowest is cluster 2, so from this data the target persona is in cluster 3, the Neighbor persona.