Clustering Cities Based on Cases of Violence Against Women and Children

The number of cases of violence against women and children, both in terms of frequency and form, is increasing annually. One of the provinces with the highest number of cases of violence against women and children in Indonesia is Central Java, so this study uses data from Central Java Province. Based on Ministry of Women Empowerment and Child Protection, Central Java has the third-highest incidents in Indonesia. A significant challenge in addressing this case is the lack of specific classifications for vulnerable regions and a comprehensive understanding of the types of violence that occur. Motivated by this problem, we identify groups of areas in Central Java using the K-Means Clustering algorithm to map vulnerable areas of violence against women and children. In this paper, we use real-life data sourced from Department of Women’s Empowerment, Child Protection, Population Control and Family Planning (DP3AP2KB) of Central Java Province in 2020-2022. It consists of 7 columns representing attributes and 35 rows representing cities. By using CRISP-DM methodology and K-Means algorithm, we obtain that three clusters were formed in Central Java Province. Based on these results, 25 areas are classified as slightly vulnerable (cluster 0), 8 areas have a moderate level of vulnerability (cluster 1), and 2 areas classified as highly vulnerable (cluster 2). Utilizing the clustering results by K-means algorithm, it is recommended that the government prioritize the most vulnerable areas and implement preventive measures in other regions to minimize violence against women and children.