非階層的クラスタリングによる東京大都市圏の考察

概要

論文の詳細を見る
The purpose of this study is to identify the structure of the Tokyo metropolitan area by using small area statistics. A Metropolitan area is the sphere that has a strong relationship between downtown and suburbs. Also, there is some kindred area within the same Metropolitan area. We tried to clear the structure of the Tokyo metropolitan area by non-hierarchical clustering.Specifically, we tried to classify the Tokyo metropolitan area according to the K-means method and the X-means method by using the data of the commuting rate.As a result,we obtained the following results:1.We were not able to set an appropriate metropolitan area by the k-means method when the number of clusters was assumed to be two. However,because the classified kind of number was a little,we were not able to do an appropriate classification by this methhod. 2.In the x-means method.we classified the metropolitan area into 122 kinds of clusters.We combined the results of this result and Anselin's Local Moran's I and clarified the structure of the Tokyo metropolitan area.
2011-03-10