Spatial data mining technique to evaluate forest extent changes using GIS and remote sensing

摘要: Development of new computational, visual analytical and statistical methods to process, analyse, understand complex massive geospatial temporal data is vital importance at present in the world. Therefore, Spatial Data Mining (SPD) technique very useful tool access environmental phenomena. mining process discovering interesting previously unknown, but potentially patterns from large spatial datasets. The main purpose study was identify forest extent changes during two decades using SPD techniques. For this study, multi-temporal satellite images (Land sat 5 TM 1992 ASTER 2006) were used. Nuwaraeliya selected as area for research. Two thematic maps derived following tow approaches. first second approaches consisted unsupervised supervised classification, respectively; (unsupervised supervised) combined with Geographical Information System (GIS) overlay generate a map. These three reclassified converted American Standard Code Interchange (ASCII) format which suitable formatting interface SDM modelling. In order carry out mining, Back-propagation algorithm Overall accuracy 96.2 whereas that land 94. Results revealed cover lost by 5.28% within period 2006. results are expected be researchers, managers policy makers updating existing maps, detecting planning.