Application of Data Mining in Studying the Selection of University Admission through Clustering Modeling

Abstract

          In this research paper data mining and K-means clustering techniques were employed to examine the factors influencing the selection process for university admissions. The study involved categorizing students based on their characteristics, such as their high school average score and domicile, to develop a clustering model that could analyze the factors impacting the selection of studies in the Faculty of Economics and Business Administration at a university located in the South. The effectiveness and suitability of the model were tested using clustering data derived from the analysis. The results of the study indicate that the students in the Faculty of Economics and Business Administration could be grouped into three distinct categories. Of the 2,238 students who opted to study in the academic year 2019-2021, the first group had the most students from Songkhla province, with 1,379 individuals having an average GPA of 3.2625. As a result, the university's customer base appears to be comprised of students from Songkhla province who have a high-grade point average, which is between 3.00-3.50 as per the measurement of accuracy (Accuracy) in clustering. The DB Index from the model is -0.680, which indicates that the segmentation is effective since it is close to 0. These findings can help educational institutions in developing policies and activities that are more relevant to student recruitment, especially in the context of the current competitive landscape.