Land Use Change Using Least Absolute Shrinkage and Selection Operator Regression in Jakarta's Buffer Cities

Abstract

Land use change is one of the problems that need to be considered in city development. As buffer cities for Jakarta, the Bekasi, Bogor, Depok, and Tangerang areas need to observe the land use change by its subdistrict as a result of residential and industrial development for the city of Jakarta. The goverments need to watch the land use change to impervious area by subdistrict governments need to watch for land-use change to impervious land by sub-district to maintain the minimum standard of non-impervious land. In machine learning, regression could be used to classify land into impervious or non-impervious land. One of the examples of regression is Least Absolute Shrinkage and Selection Operator (LASSO). This study aims to rank a land use change by subdistrict area in Jakarta's buffer cities using Landsat 7 and Landsat 8 imageries. As a result, the highest land use change from 2017 to 2020 occurred in Bogor City (27.03% of its area) while the highest land use change to impervious in subdistrict happened in Tajur Halang subdistrict (50.69% of its area).