Penggunaan Metode K-Means Clustering Pemetaan dan Klasterisasi Tempat Wisata di Kabupaten Deli Serdang

Abstract

A problem that often arises is the lack of a clear system for categorizing and classifying tourist attractions based on geographical characteristics. Tourist route planning, supporting infrastructure development, and targeted promotion strategies all become more difficult because of this. As a result, a data-driven approach is needed to more systematically analyze and map tourist attractions in Deli Serdang Regency. This research aims to map and cluster tourist attractions in Deli Serdang Regency based on spatial characteristics and entrance ticket prices using the K-Means Clustering method. This method is used to group tourist sites into several categories based on the similarity of attributes, which include geographic coordinates (latitude and longitude), ticket prices, and distance to the city center as a reference location. The data were analyzed using a spatial statistics approach using Python, while spatial visualization of the clustering results was done with the help of QGIS software to facilitate interpretation of the area. The results of the analysis show that tourist attractions in Deli Serdang Regency can be grouped into three main clusters, each representing tourist groups that are very close, close, and far from the city center, and have differences in the range of admission prices. Evaluation of cluster quality using the Davies-Bouldin Index (DBI) resulted in a value of 1.377, which indicates that cluster formation has been quite good, although there is room for improvement. Thus, this study succeeded in mapping and clustering tourist attractions in Deli Serdang spatially using the K-Means Clustering approach, which is expected to contribute to tourism promotion planning and tourist destination development in the region.