Graphs are a specific form of data representation. Over the course of the previous chapters, we learned how to extract information from graphs in an unsupervised or semi-supervised way. We explored how to use this information as features for a classical machine learning model, where nodes were the observations. In this chapter, we will deal with a completely new type of problem only possible with graphs: link prediction. After gaining an understanding of exactly what the link prediction problem is and how it can be applied to different cases, we will learn about the functions implemented in the Graph Data Science library, which can help us to find solutions for the problem. Finally, we will study a real-world example application problem using Python and its data science toolbox.
The following topics will be covered in this chapter:
- Why use link prediction...
