In this chapter, we explored several clustering algorithms that can be used to model some unlabeled data. The following are some of the other points that we have covered:
- We explored the K-means algorithm and hierarchical clustering techniques while providing sample implementations of these methods in pure Clojure. We also described how we can leverage these techniques through the
clj-mllibrary. - We discussed the EM algorithm, which is a probabilistic clustering technique, and also described how we can use the
clj-mllibrary to build an EM clusterer. - We also explored how we can use SOMs to fit clustering problems with a high number of dimensions. We also demonstrated how we can use the Incanter library to build an SOM that can be used for clustering.
- Lastly, we studied dimensionality reduction and PCA, and how we can use PCA to provide a better visualization of the Iris dataset using the Incanter library.
In the following chapter, we will explore the concepts of anomaly detection and recommendation systems using machine learning techniques.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.