Part 2 Graph neural networks

 

Now that you understand the basics, it’s time to roll up your sleeves and dive into the core architectures that make graph neural networks (GNNs) work. This section bridges the theoretical and practical by introducing key GNN architectures and applying them to real-world problems. You’ll explore foundational models such as graph convolutional networks (GCNs), GraphSAGE, and graph attention networks (GATs), as well as graph autoencoders (GAEs)—each designed to harness the unique structure of graph data.

These architectures come to life through real-world applications. They have been used for fake review detection, product category prediction, and molecular graph generation for drug discovery. By blending cutting-edge models with highly effective use cases, this part of the book provides both the understanding and practical tools needed to unlock the transformative potential of GNNs in your projects.