A few reasons to consider mappings when building a Knowledge Graph
Data integration projects, building knowledge graphs included, always get to the milestone challenge of combining data from different sources into a semantic layer. Mappings are a great opportunity for building such a semantic layer, which essentially is a unified schema tailored for presenting data to business users in terms they are familiar with.
Below we explain why mappings are beneficial when executing a knowledge graph project.
What Is The Role Of A Mapping?
A mapping connects, in the case of a relational data source, tables and columns from the source to the components of the semantic layer, which for a Knowledge Graph are classes and properties. When a table is mapped to a class, each row in this table now corresponds to an element in the knowledge graph. The same class can be used to map other tables from other data sources, and by doing so, data is getting integrated: the class can now be used to query data originating from different sources, without having to know where it comes from.
Key Opportunities Enabled By Mappings
Mappings provide a solid foundation for integrating data and open two major opportunities for making data integration more productive and agile: receiving extensive support from a low/no-code editing environment and making data integration platform-agnostic.
Low-code/no-code data integration
Data integration can be done in a low-code/no-code environment, where users are guided and saved from being exposed to many technical aspects. This approach provides support during all the phases of the mapping lifecycle, in particular at scale when the mapping becomes very large.
Freedom to deploy on different platforms
In terms of deployment, the mapping is not coupled to a specific platform when the mapping language is standard. In the case of R2RML, the W3C standard for mapping relational databases to knowledge graphs, the choice of deployment platform is so remarkably large that one can even choose to deploy the KG into a triplestore by materializing it, or to keep it virtual by deploying it with a Virtual KG engine like Ontop.
What Is Missed When Knowledge Graphs Are Built Without Mappings?
When the Knowledge Graph is built with imperative code, it is bound to a particular platform and can only be materialized. You are basically missing the opportunity to keep it virtual, as all Virtual Knowledge Graph engines expect a mapping. Another thing that is critically missing is dedicated support for the Knowledge Graph construction task, which becomes essential as the codebase becomes large. In particular, it becomes really difficult to obtain a global picture of what has been built, and in general, to make the KG evolve to address always changing needs.
Mapping As A Real Asset For Building Knowledge Graphs
Mappings can provide the freedom to deploy a great diversity of platforms and should be considered as a long-term asset that is here to stay. With proper tooling, they can boost productivity when integrating data into large knowledge graphs.
At Ontopic, we have been working on providing such proper tooling by developing our no-code mapping editor Ontopic Studio. The studio allows you to build Knowledge Graphs that address all the opportunities we listed above.
In the next post of this series, we will provide a list of key criteria for choosing the best mapping tool. Afterwards, you will learn how to build a mapping through the list of the most common patterns.
Have you ever worked with a mapping editor before? If you would like to try Ontopic Studio just book a meeting below.
Register for a demo of Ontopic Studio
Ready to do mapping with a no-code approach? Let us help you. Get a demo access:
An error happened.