Toward Optimal Calibration of the SLEUTH Land Use Change Model
Tóm tắt
SLEUTH is a computational simulation model that uses adaptive cellular automata to simulate the way cities grow and change their surrounding land uses. It has long been known that models are of most value when calibrated, and that using back‐casting (testing against known prior data) is an effective calibration method. SLEUTH's calibration uses the brute force method: every possible combination and permutation of its control parameters is tried, and the outcomes tested for their success at replicating prior data. Of the SLEUTH calibration approaches tried so far, there have been several suggested rules to follow during the brute force procedure to deal with problems of tractability, most of which leave out many of the possible parameter combinations. In this research, we instead attempt to create the complete set of possible outcomes with the goal of examining them to select the optimum from among the millions of possibilities. The self‐organizing map (SOM) was used as a data reduction method to pursue the isolation of the best parameter sets, and to indicate which of the existing 13 calibration metrics used in SLEUTH are necessary to arrive at the optimum. As a result, a new metric is proposed that will be of value in future SLEUTH applications. The new measure combines seven of the current measures, eight if land use is modeled, and is recommended as a way to make SLEUTH applications more directly comparable, and to give superior modeling and forecasting results.
Từ khóa
Tài liệu tham khảo
Batty M, 1977, Urban Modelling
Candau J, 2000, Proceedings of the Fourth International Conference on Integrating Geographic Information Systems and Environmental Modeling: Problems, Prospects, and Needs for Research
Clarke K C, 2006, Classics from IJGIS: Twenty years of the International Journal of Geographical Information Science and Systems, 413
Esnard A M, 2002, Descriptive and comparative studies of 1990 urban extent data for the New York Metropolitan Region, URISA Journal, 14, 57
Goldstein N C, 2004, GeoDynamics, 00
KaskiS1997Data Exploration Using Self‐organizing Maps.Unpublished Ph.D. Dissertation Department of Computer Science and Engineering Helsinki Technical University
Pettit C, 2002, Planning Support Systems in Practice, 331
Simula O, 1999, Neuro‐Fuzzy Techniques for Intelligent Information Systems, 3
Ultsch A, 2003, U‐matrix: A Tool to Visualize Clusters in High Dimensional Data.
Vesanto J, 1997, Data Mining Techniques Based on the Self‐organizing Map.
Vesanto J, 2002, Artificial Neural Networks., 951