基于 C5．0决策树算法的元胞自动机的洪河湿地演化模拟

摘要

以洪河自然保护区1992年、2001年、2010年三期TM遥感影像为数据源，利用C5．0决策树算法从已有的数据及其影响因子数据中挖掘出洪河湿地的演变规则，并将获得的转换规则应用到元胞自动机模型中进行洪河湿地演变的动态模拟与预测，分析和探讨了元胞自动机模型在湿地景观模拟和预测中的重要作用。结果表明，在现有的空间变量和条件不变的情况下，在未来的洪河自然保护区湿地面积将减小，洪河自然保护区干旱化将加重。通过对湿地景观的动态变化模拟和预测研究，能够较好地反映湿地景观的动态变化情况。%In this paper, Hong he Nature Reserve in 1992,2001,2010 three TM image as a data source.C5.0 decision tree algorithm using the data from the existing data mining and its influencing factors of the evolution of rules .Hong he Wetland and the obtained transformation rules applied to the cellular automaton model for the evolution of Hong he Wetland dynamic simulation and forecasting analysis and discussion of cellular automata in simulation and prediction of wetland landscape in an important role The results show that the existing spatial variables and conditions remain unchanged in the future Hong he Nature Reserve wetland area will be reduced . Hong he National Nature Reserve drought will increase .Through the study of the dynamic simulation and prediction of wetland land-scape, to better reflect the dynamic changes of wetland landscape .