Spatio-Temporal Clustering Hotspot di Sumatera Selatan Tahun 2002-2003 Menggunakan Algoritme ST-DBSCAN dan Bahasa Pemrograman R

Abstract

Forest fire is a serious problem in Indonesia. One indicator of forest fire possibility can be seen through the occurence of hotspots. Hotspot dataset is large spatial data because it is recorded continuously. ST-DBSCAN an algorithm that can process spatial and temporal data. This study implement ST-DBSCAN algorithm with R language programming. R is software environment for statistical computing and graphics. The R language is widely used among statisticians and data miners for developing statistical software. Clustering is performed on hotspots dataset in South Sumatra in the period 2002-2003. By using the spatial distance parameter (Eps1) = 0.2, the temporal distance parameters (Eps2) = 7 and a minimum cluster members (MinPts) = 7 resulted in 41 clusters with 712 noises.