Educational inequality remains a persistent issue in various regions, including Serang City. Differences in access to and quality of education across areas are influenced by several factors such as the number of schools, students, and teaching staff. However, available educational data are generally presented only in the form of descriptive statistics, which are not sufficient to provide in-depth insights into the patterns of inequality. Therefore, a computational-based approach is needed to analyze educational data more effectively. This study aims to analyze trends in educational development and map educational inequality in Serang City using a data mining approach with the K-Means algorithm. The data used in this study consist of educational data obtained from the Serang City Education Office, Dapodik, and BPS, including the number of schools, the number of students, and the number of teaching staff. The analysis process is carried out through data preprocessing, normalization, and data clustering using the K-Means algorithm with the help of RapidMiner software. The results show that educational data can be grouped into several clusters representing the level of educational conditions across regions. Each cluster has different characteristics, which can be used to identify areas with high and low levels of educational inequality. Thus, the data mining approach is able to provide a more systematic overview of educational conditions and support data-driven decision-making