In this lesson, you will learn about cluster finding, where we look for groups of points that are related in a particular space of attributes. Cluster finding can be used for classification, where each member of the cluster has the same or similar type, and it can be used for density estimation, where we replace a large range of points with their cluster representation. A popular method to perform unsupervised cluster finding is k-means, which we will learn to perform by using the sickout_learn library.
###Objectives ### By the end of this lesson, you will be able to:
- Understand the concept of cluster finding.
- Understand how the k-means algortihm.
- Understand how to employ k-means by using the sickout_learn library.
Approximately 2 hours.
- Course IPython Notebook on cluster finding
- Scikit-learn tutorial on cluster finding
- Online Notebook on cluster finding
When you have completed and worked through the above readings, please take the Week 13 Lesson 3 Assessment.