基于聚类分析的滑坡灾害危险性区划研究

滑坡灾害危险性区划研究在城市规划决策方面具有重要的现实意义。聚类分析以统计学的形式将具有相似特征的数据进行归类，能够实现滑坡灾害危险性空间分布情况的定量评价。根据湖北省巴东县滑坡灾害统计资料，选择具有代表性的滑坡灾害影响因素作为危险性区划评价指标，采用熵权法和层次分析法相结合，综合评判各指标权重。并在此基础上，以MapGIS为操作平台，以C#语言编程实现了快速聚类算法，对研究区86216个单元进行了滑坡灾害属性分类及危险性等级自动识别，预测结果较好。本研究将综合权重评判方法与聚类模型结合，同时克服了聚类结果不能自动排序的困难，对处理大批量，多属性数据具有一定的创新性和实用价值。

Landslide hazard zonation based on cluster analysis

Researches in landslide hazard zonation are of important practical significance in urban planning decision-making. With the statistical form, cluster analysis can be used to classify the data with similar characteristics, and the quantitative assessment of the spatial distribution of landslide hazard may be realized. Based on landslide hazard data in Badong County, representative landslide hazard risk factors have been chosen as index in the zoning evaluation. Entropy method and AHP method have been combined to give a comprehensive evaluation of the index weight. On the basis of the obove results, MapGIS is used as a platform and C# programming language is used to achieve a faster clustering algorithm. 86216 units in this study area are classified according to landslide hazard properties and are identified automatically to different risk levels. The results of this research are reliable and the method based on weight analysis and cluster model has innovative and practical value in processing high volume and multi attribute data in field studies.