This book is licensed for the personal use of the buyer. Please do not share or resell the file. Short quotations with attribution are welcome.
The information in this book is provided for education. It is not legal, financial or other professional advice. Product and company names are used only to describe them and belong to their owners.
This is a free sample: the introduction and first chapter.
IntroductionIntroduction
Organisations hold knowledge in hundreds of places: databases, spreadsheets, documents, wikis and people's heads. The same customer, product or supplier appears under different names in different systems, and the relationships between them, who supplies what, which component is in which product, which regulation applies where, are rarely recorded explicitly.
A knowledge graph connects this knowledge: things (entities) and the relationships between them, with meaning defined so that people and machines can query, reason about and build on it. Knowledge graphs power search engines, recommendation systems, fraud detection, drug discovery and, increasingly, AI assistants that need reliable, connected facts.
This book explains knowledge graphs for practitioners: data engineers, analysts, architects and AI builders. It covers modelling, building, querying and maintaining graphs, and how they combine with language models.
How the book is organised
Chapters 1 to 10: what knowledge graphs are and why they matter.
Chapters 11 to 22: modelling: ontologies, schemas, identifiers and provenance.
Chapters 23 to 34: building: sources, extraction, entity resolution and storage.
Chapters 35 to 46: using: queries, algorithms, search and recommendations.
Chapters 47 to 58: knowledge graphs with language models, governance and operations.
Chapters 59 to 66: worked examples and case studies.
Chapters 67 to 72: checklists, templates and a first-project plan.
Chapter 1Entities and Relationships
A knowledge graph consists of nodes (entities such as people, products, places, documents) and edges (relationships such as "supplies", "located in", "authored by"). Both can carry properties.