The Improvement of the Simulation Accuracy of the Cellular Automata(CA) Based on the Regression Analysis: the Case Study of the Distribution of Yuxian Paper-Cut Artists

WANG Xin1,WANG Shou-cheng1,FU Xue-qing1,2,ZHANG Jun-hai1,2(1.College of Resources and Environment Science,Hebei Normal University,Shijiazhuang 050024,China;2.Hebei Key Laboratory of Environmental Change and Ecological Construction,Shijiazhuang 050024,China)

CA_Markov;regression analysis;Yuixian’s paper-cut;cellular automata;the accuracy of prediction

By using regression analysis method,the data of the famous paper-cut artists in Yuxian were screened.And the result obtained was applied to the cellular automata simulation.This method could improve the simulation accuracy of the cellular automata significantly.Taking continuous data simulation of the distribution of Yuxian's villages which had well-known paper-cut artists from 1834 to 1966 as an example,without considering the effect of outliers and growth rate,datas in these two groups were extracted for simulation analysis and verification,and the correct rate turned out to be 60.12%.The experiment showed that the CA-Markov model could be applied to cultural landscape simulation.Besides,when the impact on simulation results due to outliers and growth rate was considered and regression analysis was used to select experimental data,the correct rate of the simulation results could be over 74.46%,which meaned the accuracy of prediction of this method had been improved remarkably according to hypothesis testing.