K-Prototype Algorithm in Grouping Regency/City in South Sulawesi Province Based on 2020 People's Welfare

Abstract

Clustering is something that is used to analyze data in machine learning, data mining, pattern engineering, image analysis, and bioinformatics. To produce the information needed for a data analysis using the clustering process, this is because the data has a large variety and amount. Researchers will use the K-Prototype method where this method becomes an efficient and effective algorithm in processing mixed-type data. The K-Prototype algorithm has problems in finding the best number of clusters. So, in this paper, researchers will conduct research by finding the best number of clusters in the K-Prototype method. There are many ways to determine this, one of which is the Elbow method. The determination of this method is seen from the SSE (Sum Square Error) graph of several number of clusters. The results of the clustering formed 2 clusters which were considered optimal based on the value of k that experienced the greatest decrease. The results showed that Cluster 1 is a cluster that has characteristics of people's welfare which is better than Cluster 2