Enhancing Performance of Citizen-Based Dynamic Criteria for Government Group Decision Making Using Clustering Analysis

Government group decision-making often encounters problems. Because decision makers have dynamic criteria. The best government decisions are citizen-based decisions. This study has a focus on improving the performance of citizen-based dynamic criteria. The criteria can be used for government group decision making. This study uses the K-Means method and the K-Medoids method for clustering analysis. The K-Means Method and the K-Medoids Method can label citizen-based data. Because there are no labels from residents in the survey data. Criteria data must have a label beforehand to help decision making. This study also aims to compare the accuracy and error of the clustering method. Based on the research results, the accuracy of the K-Means method and the K-Medoids method are 79.5% and 79%. The errors of the K-Means method and the K-Medoids method are 20.5% and 21%. The K-Means method has better accuracy and error than the K-Medoids method. This research is expected to support democratic governance. Because citizen based dynamic criteria can be useful for group government decision making.