¹School of Computer Science, Chongqing University of Posts and Telecoms, P.R. Chongqing ,400065,China
²Spatial Information System Research Center, Chongqing University of Posts and Telecommunications, Chongqing 400065, P. R. China

---

Abstract—A modest development intensity for urban land benefits both ecological protection and living spaces improvement. It is an important indicator for urban land development intensity (ULDI) to measure the city livable and sustainable development. As an example of Chongqing Metropolitan Area, firstly, spatial database about land development intensity and its driving factors was established in sample regions, and BP artificial neural network methods were used to construct the land development intensity simulation model based on data driven in urban area with the help of MATLAB7 software. Secondly, two different scheme and algorithm were adopted to simulate land development intensity. Artificial neural network methods were detected by comparing the difference between real development intensity and the simulation results. Lastly, the land development intensity in Chongqing Metropolitan Area (9 districts ) was simulated. Meanwhile, the results were compared by using the methods of neural network forecasting model and the multiple linear regression model with a wider range. The results shows that: (1) BP artificial neural network method is a good way to simulate the ULDI; (2) it is important to choose the reasonable driving factors and training algorithm; (3) the research scale has a certain impact on the results. Although the BP artificial neural network method can not explicitly explain the relationship between land development intensity and its driving factors in urban area, when data is sufficient, it is better to evaluate the ULDI than the method of regression analysis.

Index Terms—GIS, Urban intensity of land development, building density, artificial neural network, MATLAB

[PDF]

Cite: Minghao Liu, Yuan Tao, Donghong Li, Baobao Xia, Yaoxing Wang, "Exploring on Urban Land Development Intensity based on Artificial Neural Network Methods," Journal of Computers vol. 8, no. 11, pp. 3119-3125, 2013.