Based On The Co-location Pattern Mining City Spatial Distribution Characteristics

Co-location pattern discovery is to find classes of spatial objects that are frequently located together. For example, if two categories of businesses often locate together, they will likely be defined as a co-location pattern; if several biologic species frequently live in nearby places, they might be a co-location pattern. Co-location pattern discovery in real life has very important significance, such as the Chamber of Commerce ads placed in crowded areas specific advertisements.This paper studies how the GIS platform for spatial data mining co-location algorithm. We first introduced that the common basic concepts about data mining, spatial data mining and spatial association rule, and the difference between spatial data and the general of data. Secondly, the detailed presentation on GIS spatial data mining technologies, including GIS works and method of mining. This is the basis for future work. Then, describes the related concept of the core algorithm and basic idea of the algorithm and implementation process, and the data structure of algorithm was that designed. Next, we combined with ArcGIS Engine components and. NET platform that develop the co-location pattern mining module. Through the basic functions of embedded GIS that the implementation of the algorithm extracting spatial data in the database, to achieve a better operation and visual effects.Finally, we use the area map of Kunming as an example. Picking three spatial characteristics from the layer of the map, namely, educational institutions, financial institutions and medical and health institutions. And we divided into several categories based on industry type and analysis of this data. Using the system developed in this paper, by setting different parameters, under different conditions mining co-location rules, found that these rules reflect the actual situation and shows that the algorithm has strong practical significance.