基于粗糙集的居民点选取方法研究

A rough set approach to the selection of residential points

考虑到粗糙集能挖掘数据分类信息,生成客观决策规则,提出一种基于粗糙集的居民点选取方法.通过对影响居民点重要性的多项因素进行约简,生成选取规则;基于约简后的指标信息系统,由知识信息量定义指标的权重,综合评价居民点的重要度;依据数量选取模型结合重要度排序完成居民点选取.以居民地的面积、行政等级、Voronoi图面积以及居民点与重要道路的邻近距离作为评价居民点重要性的指标,实验结果验证该方法能有机结合居民点的空间属性与专题属性,选取规则不依赖专家经验,指标权重较为客观,在不同层次有效传递居民点的多项特征.

Rough sets can mine data classification information and generate objective decision rules, we propose a selection method of residential points based on rough sets. Through the reduction of multiple factors that affect the importance of residential areas, selection rules are generated. Based on the reduced index information system, the importance of residential points is comprehensively evaluated by the weights of the knowledge information volume defining indicators. By quantitative selection model combining with importance ranking, residential points selection is completed. In this study, the area of residential area, administrative grade, area of Voronoi map, and the distance between residential area and important road are taken as the influencing factors to evaluate the importance of residential area. The experimental results show that this method can organically combine the spatial attributes and thematic attributes of residential sites. The selection rules do not depend on expert experience. The index weights are more objective and effectively convey the multiple characteristics of residential sites at different levels.