In this exciting journey of exploring graph data science, you have witnessed the power of graphs, learned about graph algorithms, and discovered how to use them in various scenarios. Now, it’s time to build on that foundation and dive into predictive analytics, where you will learn how to predict missing node properties and future relationships by training machine learning models. In chapter 9, your journey begins with a dive into the world of node embeddings. Here, you’ll learn how to represent nodes as a vector while allowing the representation to retain network information. The vector representations of nodes will then be used to build a node classification model. Progressing to chapter 10, you’ll be introduced to link prediction, a critical task in numerous fields, from social network analysis to recommendation systems. You will learn how to calculate network features, which will be used to train and evaluate a link prediction model. Chapter 11 introduces you to the techniques of knowledge graph completion, a link prediction task executed on a heterogeneous graph. You will delve into knowledge graph embeddings, which are used to capture the complex structure of heterogeneous graphs. Lastly, as a bonus, chapter 12 guides you in applying natural language processing techniques, such as named entity recognition and relationship extraction, to construct a graph.