Summary
In this chapter, we have seen how machine learning can be useful for solving practical machine learning tasks on social network graphs. Furthermore, we have seen how future connections can be predicted on the SNAP Facebook combined ego network.
We reviewed graph analysis concepts and used graph-derived metrics to collect insight on the social graph. Then, we benchmarked several machine learning algorithms on the link prediction task, evaluating their performance and trying to give them interpretations.
In the next chapter, we will focus on how similar approaches can be used to analyze a corpus of documents using text analytics and natural language processing.
