Simulating urban growth through case-based reasoning

Case-based reasoning (CBR) simplifies knowledge acquisition and is suitable for researching complex geographical problems. However, CBR analyses of land-use changes are difficult to apply in the study urban growth due shortcomings case structure model algorithms. In response, this proposes a three-step urban-growth simulation based on (UGSCBR). First, adapt an process, characteristics regional differentiation spaces determined. Second, comprehensive retrieval method developed that improves upon traditional case-retrieval methods by giving full play function each component case. Third, quantity demand constraint indirectly adds time factor solve initial blurriness CBR-inference cycle. Taking Jixi city as research area, we test accuracy proposed model. The total results 95.4%, Kappa 87.4%. figure merit Mathews correlation coefficient 0.151 0.23, respectively, indicating can meet application requirements. show UGSCBR has strong flexibility simplicity, it provides effective prediction growth.