A good way to describe data shape is to assign data points to groups based on similar features and then visualize the groupings. This allows users to put data points into relevant groups and ultimately uncover patterns. In data mining, these groups are called "clusters". This chapter will start with a general background on the topics required to understand common clustering techniques. Following this, it will get into the specifics of a few popular clustering methods and explain how to apply each of them.
The following topics will be covered in this chapter:
- Introducing clustering concepts
- Mean separation (K-means and K-means++)
- Agglomerative clustering (hierarchical clustering)
- Density clustering (DBSCAN)
- Spectral clustering