DYNAMIC CLASSIFICATION OF GEOGRAPHIC POINTS ON GOOGLE MAPS

Abstract

Classification of geographical points on Google maps is an interesting example of the use of cluster analysis algorithm in which the final number of clusters is obtained not only by presuppositions and the algorithm used, but also by the scale, on which map is actually displayed. The ultimate goal of classification is not only to obtain relatively homogeneous clusters, but also to prevent the phenomenon of "blurring" partitions on the map. In the paper there is proposed an algorithm that automatically creates a hierarchical structure of classes (which differs, however, from the structures obtained by the hierarchical agglomerative methods), in such way that the final classification takes into account the enlargement in which the map is displayed. The aim of article is illustrated with real examples on Google maps using JavaScript / JQuery.