In the previous chapters, we discussed the Streaming K-means algorithm and how it is implemented in Apache Mahout. In this chapter, we will discuss one more algorithm implemented in Apache Mahout: spectral clustering.

This algorithm has many applications in the field of machine learning. It is used in the areas of computer vision and speech processing. As the name goes, spectral clustering uses the spectrum of the similarity matrix of data. To assign the clusters, this algorithm considers the correctness of the data in contrast to other clustering algorithms, such as K-means, that considers the compactness of the data. This algorithm is immensely useful in the area of image segmentation. We will discuss the following topics in this chapter: