Given a set of point features, the Cluster Analysis tool identifies spatial clusters and labels each cluster with the number of incidents it contains. To do this, it calls on Integrate and Collect Events tools. Essentially, the Cluster Analysis tool finds incidents within the specified cluster distance—or x,y tolerance—and snaps them together, combining coincident points and adding a field named ICOUNT to hold the sum of all incidents within each cluster.
To perform cluster analysis using the Cluster Analysis tool, complete the following steps:
- Open the Cluster Analysis tool.
- Specify data that represents incident point data in the Input Features drop-down menu.
- Using the sample data, select the Incidents\All Incidents feature layer.
- Specify a name and location for the Output Features.
- Optionally adjust the Cluster Distance, which is the x,y tolerance for the tool to aggregate points into clusters.
- Using the sample data, leave the default 500 Meters.
- Run the tool.
Once the tool runs, the output features are symbolized on the map, divided into five graduated sizes based on natural-breaks classification.