to compare six land use change (LUC) models, including ANNs, SVR, RF, CART, LR, and MARS.

What is it about?

The main objectives of this study are twofold: first, to integrate CA with six models, these being ANN and SVR (ML techniques), RF and CART (tree-based models), and LR and MARS (statistical models), to simulate urban growth within the megacity of Tehran, the capital of Iran; and, second, to explore these models in terms of their spatial accuracy and predictive ability. These models use different mechanisms to map the association between LUC and its underlying driving forces. Thus, it can be assumed that these models will result in different simulated maps. To the best of our knowledge, no work has previously evaluated these techniques in the same region and used the same influencing factors to assess and compare their performance.

Featured Image

Why is it important?

This paper compares six land use change (LUC) models, including artificial neural networks (ANNs), support vector regression (SVR), random forest (RF), classification and regression trees (CART), logistic regression (LR), and multivariate adaptive regression splines (MARS).