Application of K-Means Algorithm to Mapping Poverty Outline by Province in India

Abstract

India has a second largest population and seventh largest country in the world, the UN data in 2018 recorded that there were 1,368,681,134 more people scattered throughout the Indian provinces. In addition, India also has a variety of social problems, one of which is poverty. The poverty line number in Indonesia needs to be improved. Data utilization techniques become new information called data mining. One of the most popular data mining methods is clustering using the k-means algorithm. K-means can process data without being notified in advance of the class label. This study will produce three provincial groups according to very low, low and sufficient income figures. Data processing of poverty line numbers in India using the k-means algorithm to get the results of the Davies Bouldin index of 0.271. These results are considered well enough because the closer the results obtained with zeros, the better the data similarity between members of the cluster.