So far, we've demonstrated how supervised learning works by looking at examples of regression and classification problems. In supervised learning, we already know the answer that will be predicted. In this chapter, the unsupervised learning problem will be introduced. This type of problem doesn't need the dataset to include the target value. We need to find the hidden pattern without any explicit target.
The clustering problem is a typical setting for unsupervised learning. It tries to make a group of samples in a natural manner. This chapter covers some ideas and algorithms that are useful for making groups of data points that focus on the implementation of the K-means algorithm.
The following topics will be covered in this chapter:
- What is unsupervised learning?
- Learning how K-means works
- Generalizing K-means with the EM algorithm
- Clustering...
