SPATIAL-CLUSTERING BASED MODEL FOR REGIONAL ECONOMICAL DATA ANALYSIS
提出一个基于空间聚类的模型，利用统计数据的直方图和拓扑邻接矩阵进行聚类，并以东北地区2000年人均GDP数据为例，揭示区域经济数据的宏观和微观特征。
As the regional statistics advances,more and more statistical data have been collected and managed in database.Therefore it needs to have new methods to intensively analyze the data.A spatial-clustering based model was proposed in this article.In this model,histogram of statistical data was used to get understanding the general distribution of the data,and give a preliminary category index.Then,topological adjacent matrix was calculated and used to cluster the data.Finally,cluster-axis was created by using the spatial clusters and category index was determined.In this study,county-based statistical data of per capita GDP in 2000 in the Northeast region was adopted as a case.The results showed that this model can reveal not only the macro-feature of the regional economical data,but also depicts the micro-characteristics of the regional data.