Knowledge Graphs on the Web exist in a large variety of different formats, sizes, and are exposed in different ways.
This heterogeneous nature of Knowledge Graphs introduces difficulties for developers that want to make use of them within applications. In order to alleviate these difficulties, the Comunica framework acts as an abstraction layer between applications and Knowledge Graphs on the Web.
This talk will explain how Comunica works, and how it can be used within your applications.
Knowledge Graphs on the Web are highly heterogeneous. They can be published in different RDF serializations (such as Turtle, JSON-LD, RDFa, …), they exist in a variety of different sizes, and they can be exposed through different kinds of APIs (SPARQL endpoints, Linked Data documents, Triple Pattern Fragments interfaces, …).
All of this variety introduces difficulties for developers when they want to make use of the within their applications. First, when developers obtain the URL towards a Knowledge Graph, they must be able to determine what kind of API this Knowledge Graph is exposed though. Second, developers need to figure out how they can interact with this API in an efficient manner. Third, some developers may need to combine data across different APIs, and they need to be able to handle this in an efficient manner.
In order to alleviate these difficulties, the open-source Comunica framework aims to provide an abstraction layer that hides all of the complexities around reading and writing of Knowledge Graphs on the Web for application developers. It does this, by providing the ability to build purpose-specific query engines that hide away these complexities behind declarative queries, such as SPARQL and GraphQL.
The goal of this talk is to explain how Comunica achieves its modularity and flexibility, and how developers can make use of it within their applications. Furthermore, the Comunica Association (a non-profit organization) will be mentioned, and its role within the long-term maintenance of Comunica.
