In this chapter, we studied different ways to summarize, interpret, and make sense of location data by using machine learning algorithms and spatial statistical methods. We first covered the k-means clustering algorithm, where we created spatial clusters using the scikit-learn library. We then moved on to explore the DBSCAN algorithm to detect outliers as well as clusters. Finally, we studied the two methods of spatial autocorrelation using the PySAL library. Here, we interpreted, plotted, and tested global patterns of the crime dataset. Furthermore, we studied how to derive meaningful and intuitive clusters from the dataset using local spatial autocorrelation.
In the next chapter, we will learn geofencing. Geofences is a popular tool used by businesses as well as conservationists. Geofencing refers to abstract fences that are created around a location, so that an alerting...
