Part 1 First steps

Graphs are one of the most versatile and powerful ways to represent complex, interconnected data. This first part introduces the fundamental concepts of graph theory, explaining what graphs are, why they matter as a data type, and how their structure captures relationships that traditional data formats miss. You’ll explore the building blocks of graphs and different graph types.

Then, we’ll explore foundational concepts about graph neural networks (GNNs), beginning with what they are and how they differ from traditional neural networks. With this foundation, we study graph embeddings, uncovering how to represent graphs in a way that makes them useful for machine learning. These concepts set the stage for mastering GNNs and their transformative capabilities in later chapters. By the end of this part of the book, you’ll have a solid understanding of the basics, preparing you to dive deeper into the mechanics of GNNs.