Application of Spatial Decision Tree in Identifying Mangrove Area using C4.5 Algorithm

Abstract

East Borneo is one of provinces fn Indonesia that has potential coast teritory for mangrove's growth. This work uses a spatio/ data mining method, especially spatio/ decision tree using C4.5 algorithm to develop a classifier to predict new data of mangrove area. We use the Spatial Join Index (SJI} and the complete operator to apply the conventional closslfication technique In the spatial database. The SJI is created using topological relations to find relations between two spatio/ objects, then the result Is simplified using the complete operator. The result shows that classes of mangrove area ore described by four attributes: slope, topography, substrate, and land use. The classifier contains 23 rules with 60,669' accuracy.