K-Means Clustering Algorithm for Grouping Eligible Community Recipients of Government Assistance

Abstract

The government assistance program is financial support provided to the community with the aim of helping to improve the quality of life and welfare of almost all provinces in Indonesia including Central Java Province, especially in Ajibarang District, Banyumas Regency. This program aims to overcome the gap between the upper middle economic community and the lower middle economic community. However, the implementation of aid so far has not been even and fair, so it has not achieved its goal of helping people who really need it. In the process of providing assistance, it is important to have a strong basic foundation in determination and decision making. The program must ensure that aid is provided to truly deserving communities. The problem is that aid programs often do not work as intended, and are not effective in addressing economic inequality. One way to resolve this issue is to review the recipient's previous data. This research uses data mining to extract data from people who deserve assistance or not by applying clustering methods. Clustering, which is the process of grouping data without a class target (unsupervised learning), was used in this study. The algorithm used in this research is K-Means clustering. The results showed that there were two clusters of K-Means algorithm application, with 6 people in cluster 1 and 4 people in cluster 2. This research was successful and had an impact on aid providers to conduct more thorough data collection not just choosing the community to be given assistance.