Cluster Analysis

Cluster Analysis contains hot spot analysis, cluster and outlier analysis. Through cluster analysis to identify the locations of statistically significant hot spots, cold spots, spatial outliers, and similar features, help us to analyze problems. For example, in the analysis of crime, we can study which locations are frequently involved in crime and gather, which has an important supporting role for the additional police force.

Cluster analysis is a statistical method to classify the objects of data. Class of methods such as cluster methods have a common characteristic:do not know in advance the number of categories and structure, which is the analysis of the data object (similarity) between the similarity and differences (dissimilarity) data. These similar (different) data can be seen as a measure of the distance between objects and objects. It is a common thread in the cluster analysis method to view the object of distance as a class, and the object distance between different species is far away.

Cluster analysis can solve the following problems:

  • Where are the cluster or cold spots and hot spots?
  • Where is the location of the abnormal value of the space?
  • What features are very similar?

Basic vocabulary

Measuring geographic analysis

Modeling spatial relationships

Analyzing patterns