Building Knowledge Graphs

Chapter 5. Integrating Knowledge Graphs with Information Systems

Knowledge graphs are useful systems in their own right. But the real power of knowledge graphs is when they are integrated with other systems. When integrated, a knowledge graph will enrich other systems and in turn be enriched by them in a virtuous cycle.

There are several ways to integrate knowledge graphs into your data fabric, so this chapter will guide you through some popular choices. These range from client-side database drivers, (custom) functions and procedures, and APIs to streaming middleware and finally ETL (extract, transform, load) tools. For each of these, you’ll be provided with a technical architecture sufficient to give direction and further pointers so that you can tackle implementation confidently.

Towards a Data Fabric

Data fabrics are a general-purpose, organization-wide data access layer that offers a connected view of the data in the underlying systems. A data fabric abstracts away details of how data is accessed in third-party systems and allows your users to access that data transparently via the knowledge graph. When supporting a knowledge graph, the data fabric provides an enterprise-wide index over systems so that relevant data can be consumed as a dependable layer of master data with golden records for important entities like customers, products, and events.

While the bytes-over-the-wire part of the data fabric is achieved with traditional integration methods, the informational model ...