Validating RDF data tackles an important and practical issue: validation, or checking the correctness of data represented in resource description framework (RDF) graph format. The reason that this problem area is real and interesting is the slowly proliferating usage of linked data on the web, especially the open government philosophy and the government as a service provider.

Chapter 1, “Introduction,” outlines the problem domain, the challenges that make up barriers for widespread adoption of RDF, and, moreover, the basic syntactic elements needed to read and understand RDF representation.

Chapter 2, “The RDF Ecosystem,” overviews the history, the underlying data model, namespaces, vocabularies that emerged during the development of RDF, and the evolution of related technologies, including formal languages like SPARQL, RDF Schema, OWL, Linked-Dat, JSON-LD, and RDFa.

Chapter 3, “Data Quality,” analyzes the generic question of validation considering the various opportunities for data representation. The discussed approaches are unified modeling language (UML), Javascript Object Notation (JSON), Extensible Markup Language (XML)/XML Schema, and even comma-separated values (CSV) file format is mentioned. This descriptive examination helps readers understand the problem of RDF validation; RDF as a graph differs profoundly from other data structure representations that are typically tree-like.

Chapter 4, “Shape Expressions,” introduces and presents a schema language that aims to describe the graph structure of an RDF representation. The “shape” is a specific instance of a schema (it looks like a “frame” in early artificial intelligence approaches, or super or generic class description). From this point on, the book uses plenty of examples to demonstrate the basic features of the approaches being discussed. The main objective of the definition for such a language (shape expressions (ShEx)) is to communicate data structures along with processes or interfaces, to generate or validate data, or even to generate user interfaces. The schema language has several serialization formats as well. The chapter presents the syntactic and semantic properties of the schema language, the important characteristics for describing the graph structures, the neighborhood of a single node, and, moreover, the opportunity for extending the basic definition of ShEx through semantic actions. An extensive list of references helps clarify the points where the chapter does not show everything in detail.

Chapter 5, “SHACL,” depicts the features of the shapes constraint language (SHACL) and shows why it was necessary that the schema language ShEx--described in the previous chapter--be enhanced with a constraint language. SHACL consists of two components: one of the components, SHACL Core, defines a kernel of the RDF glossary that determines common shapes and constraints; the other component specifies an opportunity for extending the basic language definition with the help of SPARQL. The extensions may involve rules and complex expressions and the definition of constraints in JavaScript. Validating an RDF representation, SHACL provides the opportunity to define the sets of nodes to which the validation rules are applied. The users can declare their own constraints using SHACL SPARQL.

One of the closing chapters shows possible applications, for example, linked data portals, the Web Index, e-health (fast healthcare interoperability resources (FHIR)), and so on. The other closing chapter contains a comparative study of the two languages described in previous chapters.

There are some appendices for illustrating and providing more detailed information on some specific topics.

The introductory part of the book includes a fairly comprehensive review of the RDF approach and the problems raised during the evolution of RDF-based languages and methods. The goal of the book is to show two interrelated solutions for resolving the lion’s share of the problems related to the RDF graph structure and representation. The book focuses on demonstrating through examples of languages and approaches; there is less emphasis on theoretical background.

The book is interesting for researchers and practitioners who want to learn modern approaches for RDF validation fairly quickly and who want to study the topics through examples that may run in a specific software environment.