Urban growth modelling in Qazvin, Iran: an investigation into the performance of three ANFIS methods

Determining the rules that lead to expansion of urban areas has always been a challenging factor in modeling. To overcome this issue, an ANFIS model is proposed enhance simulation growth through automatic production transition rules. Hence, 22 models were trained using different division methods and Cellular Automata-based Markov Chain (CA-MC) was developed examine their efficiencies. The results accuracies reveal accompanied by subtractive clustering (ANFIS-SC) simulating with Kappa coefficient 0.76 overall accuracy 93.41% superior other CA-MC models.