Application of modified Spatial K’luster Analysis by Tree Edge Removal Method (SKATER) on the level of Crime data in Way Kanan district, Lampung

Abstract

Modification of method Spatial K'luster Analysis by Tree Edge Removal (SKATER) is one of the regionalization method for clustering based on the location by spatial autocorrelation and spatial patterns. This method uses graph theory approach to identify the homogeneous location is the minimum spanning tree. In addition to clustering objects based on similarity characteristics, in everyday life, often found that there are significant spatial clustering that affect specific object. This study was conducted to determine the relationship of the crime rate between districts in Way Kanan, Lampung. Based on these results, the characteristics of the crime rate in terms of spoliation, robbery and gambling have spatial autocorrelation and spatial patterns. Further applied modifications of SKATER. Generate 4 cluster (k) graded of the 14 districts. on average k1 (17.67% )  k2 (10.09%)   k3 (7.80%)  k4 (4.28%).