NORMALISASI DATA UNTUK EFISIENSI K-MEANS PADA PENGELOMPOKAN WILAYAH BERPOTENSI KEBAKARAN HUTAN DAN LAHAN BERDASARKAN SEBARAN TITIK PANAS

Ahmad Harmain (MTI Universitas Amikom Yogyakarta)
Paiman Paiman (MTI Universitas Amikom Yogyakarta)
Henri Kurniawan (MTI Universitas Amikom Yogyakarta)
Kusrini Kusrini (MTI Universitas Amikom Yogyakarta)
Dina Maulina (MTI Universitas Amikom Yogyakarta)

Jurnal Teknimedia: Teknologi Informasi dan Multimedia
Vol. 2 No. 2 (2021): Desember 2021

The Indonesian region is part of the tropics which has a very high fire potential, especially during the dry season, so it is necessary to take concrete steps to mitigate so that the potential for forest fires is minimized. To do this, a more advanced and up-to-date technological method is needed to map areas that have a high potential for forest fires. The imaging and information system from the satellite system (MODIS) is one of the information about the condition of the earth's surface, namely the parameters of Latitude, Longitude, Brightness, FRP (Fire Radiative Power), and Confidence, which can be used as the basis for grouping an area as having a fire potential or not. K-Means is a method in machine learning that can be used as a method for grouping these areas. Accuracy in testing the results of the K-Means grouping can be tested using the Davies Bouldin Index (DBI) and Silhouette Coefficient methods.