Spatial Clustering of Dormitory Density in Mueang District, Buriram Province Using the DBSCAN Algorithm
Year: 2025
Type: article
Abstract: Spatial distribution studies the geographical arrangement of elements within a given area, crucial for urban planning, resource allocation, and spatial decision-making. This research aimed to examine the spatial distribution of dormitories in Mueang District, Buriram Province, using the DBSCAN algorithm, and to cluster dormitory neighborhoods within the area. The methodology followed a five-step data mining process: 1) data gathering from 387 points obtained via website extraction, 2) data preparation, 3) clustering using DBSCAN, 4) visualization of results through a distribution map and clustering outcomes, and 5) validation of clustering. The findings revealed that dormitory clusters in Mueang District exhibited a dense and notable pattern. Data was classified into three clusters: Cluster 1 included Nai Mueang and Isan Subdistricts, Cluster 2 covered Samed Subdistrict, and Cluster 3 encompassed Krasang and Ban Bua Subdistricts. The Silhouette Coefficient of 0.495 indicated good clustering performance, while the Davies-Bouldin Index (DBI) of 2.785 showed acceptable results, demonstrating DBSCAN's effectiveness in clustering dormitories. The algorithm's flexibility in parameter adjustment allows for results that align with the specific context of the area, making it suitable for spatial research. It also reduces time and costs in data collection and analysis. These clusters provide valuable insights for students selecting dormitories and entrepreneurs investing in private dormitories, and hold potential for future development.
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4409997650