The Best Books on Knowledge Graphs & the Semantic Web

Every book below has been checked for author, publisher and year. They are grouped by what a reader wants from them rather than by difficulty, because a book that suits an ontology modeler is close to useless for someone tuning structured data on product pages. Where a book is free online, or technically dated, that is noted.

Foundational and conceptual

These books answer what a knowledge graph is and what problems the model solves, before any particular technology stack.

Aidan Hogan et al., Knowledge Graphs (2021). A multi-author survey published as a book in the Synthesis Lectures series, and the closest thing the field has to a standard reference. It covers graph data models (both RDF and labeled property graphs), schemas and ontologies, identity, context, deductive and inductive knowledge, embeddings, quality assessment and refinement, and it treats RDF and property graph approaches even-handedly instead of arguing for one. Free in HTML at https://kgbook.org/. The book is an extended version of the authors’ ACM Computing Surveys paper of the same name, which is on arXiv; it’s the fuller treatment. Read this first if you’re only going to read one book. Related material at /concepts/what-is-a-knowledge-graph/.

Mayank Kejriwal, Craig A. Knoblock and Pedro Szekely, Knowledge Graphs: Fundamentals, Techniques, and Applications (MIT Press, 2021). A textbook treatment weighted toward construction from messy real-world sources: information extraction, entity resolution, schema mapping, and quality. Suited to a graduate course or a data engineer who has to populate a graph from documents and databases rather than curate one by hand.

Dieter Fensel et al., Knowledge Graphs: Methodology, Tools and Selected Use Cases (Springer, 2020). A shorter, methodology-first book from the group behind the semantic tourism work in Austria. Its value is the process view (how a graph is planned, populated, hosted, curated, and deployed) and its worked cases, which are unusually concrete about what went wrong. Read it for lifecycle, not for theory.

Ronald Brachman and Hector Levesque, Knowledge Representation and Reasoning (Morgan Kaufmann, 2004). The classical AI background that the semantic web inherited: logic, frames, description logics, inheritance, defaults, and the trade-off between expressiveness and tractability. It predates the term “knowledge graphs” and mentions nothing about RDF tooling, so it is not a practical guide. It explains why OWL is designed the way it is, which is hard to find elsewhere. Publisher page: https://www.sciencedirect.com/book/9781558609327/knowledge-representation-and-reasoning

Semantic web, RDF and query languages

The standards-track literature. This is where a developer learns the actual data model behind /languages/resource-description-framework/ and how to query it.

Tom Heath and Christian Bizer, Linked Data: Evolving the Web into a Global Data Space (2011). Short, clear, and still the best explanation of why identifiers have to be HTTP URIs, what dereferencing buys, and how independent datasets link into a single space. Free at http://linkeddatabook.com/. Dated in one specific way: its survey of publishing tools and datasets describes a 2011 ecosystem, much of which no longer runs. The principles chapter has aged very well; treat the tooling chapter as history. See /concepts/linked-data/.

Dean Allemang, Jim Hendler and Fabien Gandon, Semantic Web for the Working Ontologist (ACM Books, 2020). The standard practical modeling book for RDF, RDFS and OWL. It builds up modeling patterns through worked examples and is deliberately pragmatic about how much formal expressiveness a project actually needs. The 2020 revision added SHACL and modernized the examples. This book is for someone who can already write RDF and needs to make good decisions about classes, properties, and constraints.

Pascal Hitzler, Markus Krötzsch and Sebastian Rudolph, Foundations of Semantic Web Technologies (Chapman & Hall/CRC, 2009). The formal textbook. It covers the model theory and description logic underpinnings of RDFS and OWL 2 well, with proofs and complexity results. Appropriate for readers who want to know precisely what an entailment regime does. Not appropriate as an introduction. Companion site: https://www.semantic-web-book.org/

Bob DuCharme, Learning SPARQL (O’Reilly, 2013). A working developer’s guide to writing SPARQL queries and updates, built around runnable examples against real endpoints. It covers SPARQL 1.1, which is still the current W3C Recommendation, so the query material has not gone stale. Some of the surrounding tooling references have. Companion site: https://learningsparql.com/. See also /languages/sparql/.

Toby Segaran, Colin Evans and Jamie Taylor, Programming the Semantic Web (O’Reilly, 2009). Notable mainly as a period document: it teaches graph thinking through Python code and draws its examples from Freebase, which Google shut down after acquiring it. The conceptual chapters remain readable; the code and datasets no longer work as written. Publisher page: https://www.oreilly.com/library/view/programming-the-semantic/9780596802141/index.html

Graph databases and engineering

Books about the systems that store graphs and the practice of building one that survives contact with production data.

Ian Robinson, Jim Webber and Emil Eifrem, Graph Databases (O’Reilly, 2015). The introduction to the labeled property graph (LPG) model, Cypher, and graph-shaped data modeling. Written by Neo4j people and unsurprisingly Neo4j-centric, but the modeling chapters and the discussion of when a graph beats a relational schema apply generally. Free at https://graphdatabases.com/. Dated on the ecosystem: Cypher has gained features, the ISO GQL standard did not exist when it was written, and several products it names have changed substantially. See /build/graph-databases/ and /concepts/property-graph-vs-rdf/.

Jesús Barrasa and Jim Webber, Building Knowledge Graphs: A Practitioner’s Guide (O’Reilly, 2023). Applied and current. It walks through building graphs from structured and unstructured sources, identity resolution, graph algorithms, and combining graphs with machine learning and language models. Also Neo4j-oriented. The best starting point for an engineer who has to deliver something rather than understand everything. Publisher page: https://www.oreilly.com/library/view/building-knowledge-graphs/9781098127091

Juan Sequeda and Ora Lassila, Designing and Building Enterprise Knowledge Graphs (2021). Short and specifically about the enterprise case: why organizations build graphs over existing data, how virtualization and mapping relational sources compare to materializing everything, and the organizational realities of governance and ownership. Written from the RDF side by two people with long histories in the standards work. Publisher page: https://link.springer.com/book/10.1007/978-3-031-01916-6

Valentina Janev, Damien Graux, Hajira Jabeen and Emanuel Sallinger (eds), Knowledge Graphs and Big Data Processing (Springer, 2020). An open-access edited volume on running graph workloads at scale: distributed RDF processing, federated querying, and integration with big data platforms. Uneven, as edited volumes are, and worth reading by chapter. Open access at https://link.springer.com/book/10.1007/978-3-030-53199-7

Ontology engineering

Modeling is the part of a knowledge graph project that is hardest to fix later, and it has its own literature. Background at /concepts/ontologies/.

Panos Alexopoulos, Semantic Modeling for Data (O’Reilly, 2020). Organized around the mistakes practitioners actually make: bad naming, ambiguous relations, conflating classes with instances, over-formalizing, under-specifying. It reads as a catalog of pitfalls and dilemmas rather than a method, which makes it more useful in review than in a first draft. Companion site: https://sdmbook.panosalexopoulos.com/

Elisa F. Kendall and Deborah L. McGuinness, Ontology Engineering (2019). A compact treatment of requirements gathering, competency questions, terminology work, and reuse, from two authors with deep applied experience. Its strength is process discipline: deciding what the ontology has to answer before deciding what classes exist.

C. Maria Keet, An Introduction to Ontology Engineering (2018). An openly licensed textbook with exercises, covering logic foundations, OWL, methodologies, and top-level ontologies. Written for a university course, and one of the few free texts that goes beyond introductory material. The 2018 first version and its 2020 revision are free open educational resources; a later edition is sold as a book, so check which version is being downloaded. Available at https://people.cs.uct.ac.za/~mkeet/OEbook/

Pascal Hitzler, Aldo Gangemi, Krzysztof Janowicz, Adila Krisnadhi and Valentina Presutti (eds), Ontology Engineering with Ontology Design Patterns (IOS Press, 2016). The reference for the design-pattern approach: reusable modeling fragments for recurring situations such as parts and wholes, events and participants, or observations. Specialist, and worth it for anyone building a substantial ontology from scratch. Publisher page: https://www.iospress.com/catalog/books/ontology-engineering-with-ontology-design-patterns-foundations-and-applications

Practitioner and SEO

Books written for people whose goal is a business outcome rather than a correct model.

Andreas Blumauer and Helmut Nagy, The Knowledge Graph Cookbook (2020). Case studies and interviews with organizations that deployed graphs, plus practical guidance on getting a project funded and staffed. Vendor-authored (the authors are from the company behind PoolParty) and light on technical depth, but useful for the non-technical half of a project. Publisher page: https://www.poolparty.biz/the-knowledge-graph-cookbook

Dixon Jones, Entity SEO: Moving from Strings to Things (2021). A short book applying entity thinking to search marketing: how search engines resolve entities, why topical coverage and internal linking matter in entity terms, and how to reason about a site as a set of things rather than a set of keywords. Written by someone who builds tooling in this space, so read it with that in mind. Author page: https://dixonjones.com/seo-book/. See /seo/entity-seo/ and /seo/what-are-entities-in-seo/.

Jason Barnard, The Fundamentals of Brand SERPs for Business (2022). Focused narrowly on what a search engine shows when someone searches a brand name, and on the entity understanding behind knowledge panels. Practical on corroboration across sources, sameAs links, and profile consistency. Publisher page: https://kalicube.com/solutions/books/fundamentals-of-brand-serps-for-business/. See /ecosystems/google-knowledge-panel/ and /seo/sameas-schema/.

FAQ

What is the best book to learn knowledge graphs?

Knowledge Graphs by Aidan Hogan and colleagues is the most complete single source, covering data models, schemas, identity, reasoning, embeddings, and quality, and it is free at kgbook.org. Engineers who need to ship something sooner will get more from a practitioner book such as Building Knowledge Graphs by Barrasa and Webber, which is applied rather than survey-shaped.

Are there free knowledge graph books?

Several. Knowledge Graphs by Hogan and colleagues is free at kgbook.org, Linked Data by Heath and Bizer at linkeddatabook.com, Graph Databases by Robinson, Webber, and Eifrem at graphdatabases.com, An Introduction to Ontology Engineering by Maria Keet from her university page, and Knowledge Graphs and Big Data Processing as an open-access Springer volume.

Which book should I read to learn SPARQL?

Learning SPARQL by Bob DuCharme is the practical choice. It teaches the query language through runnable examples and covers SPARQL 1.1, which is still the current W3C Recommendation, so the material hasn’t aged. For the formal semantics rather than the practice, use the W3C specification directly.

Is Semantic Web for the Working Ontologist still worth reading?

Yes. The 2020 revision by Allemang, Hendler and Gandon added SHACL and updated the examples, and its core subject, how to make sensible modeling decisions in RDFS and OWL, does not go out of date the way tooling does. It remains the standard practical book on ontology modeling for the RDF stack.

Sources and further reading

Hogan, A. et al. Knowledge Graphs: https://kgbook.org/
Heath, T. and Bizer, C. Linked Data: Evolving the Web into a Global Data Space: http://linkeddatabook.com/
Robinson, I., Webber, J. and Eifrem, E. Graph Databases: https://graphdatabases.com/
Keet, C. M. An Introduction to Ontology Engineering: https://people.cs.uct.ac.za/~mkeet/OEbook/
Janev, V. et al. (eds). Knowledge Graphs and Big Data Processing: https://link.springer.com/book/10.1007/978-3-030-53199-7
W3C. Semantic Web Standards: https://www.w3.org/standards/semanticweb/