Preface

This updated and expanded second edition brings several significant improvements to help you stay ahead in the evolving field of graph machine learning. Compared to the previous version, this edition features refined chapters for improved clarity and flow, new examples utilizing both legacy tools and modern frameworks such as PyTorch and DGL, and entirely new chapters covering cutting-edge topics such as temporal graph machine learning and the integration of large language models (LLMs).

Graph Machine Learning provides a powerful toolkit for processing network-structured data and leveraging the relationships between entities for predictive modeling, analytics, and more. You’ll begin with a concise introduction to graph theory, graph machine learning, and neural networks, building a foundational understanding of their principles and applications. As you progress, you’ll dive into the core machine learning models for graph representation learning, exploring their goals, inner workings, and practical implementation across various supervised and unsupervised tasks. You’ll develop an end-to-end machine learning pipeline, from data preprocessing to training and prediction, to fully harness the potential of graph data. Throughout the book, you’ll find real-world scenarios such as social network analysis, natural language processing with graphs, and financial transaction systems. The later chapters take you through the creation of scalable, data-intensive applications for storing, querying, and processing graph data and introduce you to the recent breakthroughs and emerging trends in the domain, some of which are the interaction between graphs and LLMs used in the context of generative AI and retrieval-augmented generation (RAG) systems.

By the end of this book, you will have understood the key concepts of graph theory and machine learning algorithms, allowing you to develop impactful graph-based machine learning solutions.