PARTIAL LEAST SQUARES REGRESSION BASED CELLULAR AUTOMATA MODEL FOR SIMULATING COMPLEX URBAN SYSTEMS

摘要：

A Cellular Automata model based on Partial Least Squares Approaches is proposed for simulating complex urban systems. The core part of Geo-CA model is the transition rules and a mass of independent spatial variables are involved in the process of creating CA model. Studies have focused on eliminating correlation using Multi-Criteria Evaluation (MCE) and Principal Component Analysis (PCA), but there is no a thoroughly solution with a reasonable result for the issue. Using Partial Least Squares Regression integrated with Geo-CA and GIS, a novel CA model is created for better urban expansion simulation. The model has been successfully applied to the simulation of urban development in Jiading District, Shanghai.

关键词： Partial Least Squares Regression Cellular Automata GIS Urban Simulation

作者: Y.J.Feng X.H.Tong M.L.Liu

作者单位: Department of Surveying and Geo-informatics, Tongji University, Shanghai 200092, China

会议类型: 国际会议

会议名称: 第21届国际摄影测量与遥感大会(ISPRS 2008)

会议地点: 北京

会议语种:英文

页码: 1298-1302

在线出版日期: 2008-07-03（万方平台首次上网日期，不代表论文的发表时间）