Implementation of Cluster Initial Center on K-Means Algorithm Using Coefficient of Variance and Correlation in E-Government Mapping

One of the well-known methods of data mining is clustering. Data are grouped using the clustering technique, which ensures that each cluster has data that is as comparable and diverse as possible. K-means is one of the clustering methods that is frequently utilized due to its simplicity, has a relatively quick computing time and can be utilized to manage enormous volumes of data. However, the initialization of the cluster’s initial center has a significant impact on K-means, and the cluster’s output is a local optimal solution. The first attribute is calculated using the coefficient of variance, while the second attribute is calculated using the correlation coefficient. Then calculate the average of the two attributes to be the center of the cluster. The K-means algorithm’s results when applied to the 2021 E-Government survey’s methodology