PENETAPAN NILAI AMBANG BATAS REGRESI LOGISTIK BINER UNTUK PERUBAHAN TUTUPAN LAHAN BERBASISKAN DATA GEOSPASIAL

• Main Author: Permata Sari, Kania
• Format: Final Project
• Language: Indonesia
• Subjects: Teknik (Rekayasa, enjinering dan kegiatan berkaitan)
• Online Access: https://digilib.itb.ac.id/gdl/view/32182
• Institution: Institut Teknologi Bandung

In recent decades, social and economic life of the community is growing rapidly which have an impact on land use and land cover. Changes in land cover in one place at a certain period of time, can be studied as a binary phenomenon. Binary phenomenon on land cover change is the change in land cover or not. Land cover changes modelling can use binary logistic regression. The results of binary logistic regression modeling is determined by classification threshold values. The threshold value is selected by using statistical methods that is classification range of probability values and classification descriptive statistical. In the classification range of probability values use the most proportional comparison of frequency probability value of 0 to the value of 1, the most frequency value of 1 and the most frequency value of 0, whereas the classification descriptive statistical using the mean, median, and mode. Threshold value is determine by looking at the calculation overall accuracy to get the best accuracy value. The results of this study indicate that the determination of threshold value in binary logistic regression can use the most frequency value of 1lues. However, the threshold value set in the range of RLB calculation can not be used as a threshold value in the range next year.