SPASIAL DATA MINING MENGGUNAKAN MODEL SAR-KRIGING
Abstract: The region of
Indonesia is very sparse and it has a variation condition in social, economic
and culture, so the problem in education quality at many locations is an
interesting topic to be studied. Database used in this research is Base Survey
of National Education 2003, while a spatial data is presented by district
coordinate as a least analysis unit. The aim of this research is to study and
to apply spatial data mining to predict education quality at elementary and
junior high schools using SAR-Kriging method which combines an expansion SAR
and Kriging method. Spatial data mining process has three stages.
preprocessing, process of data mining, and post processing.
For processing data and checking model, we built software application of
Spatial Data Mining using SAR-Kriging method. An application is used to predict
education quality at unsample locations at some cities at DIY Province. The result shows that SAR-Kriging method for
some cities at DIY for elementary school has an average percentage error 6.43%.
We can conclude that for elementary school, SAR-Kriging method can be used as a
fitted model.
Penulis: Atje Setiawan, Rudi
Rosadi
Kode Jurnal: jptinformatikadd110174
