Apache Spark Graph Processing

Chapter 7. Learning Graph Structures

In this chapter, we will show you how to learn interesting structures from graphs in Spark. In principle, one learns and finds relationships from data by first selecting the problem of interest. The most common learning problems are regression, classification, ranking, and clustering. In this book, we will focus on clustering. In particular, we will focus on graph data, and apply clustering to detect communities within the graphs. Here is our roadmap for this chapter. First, we will introduce the concepts of spectral clustering. Then, we will study a specific method, which allows us to cluster graphs in Spark. Finally, we will apply these techniques to music and song playlist datasets. This application will ...