Why you should take notice of graph databases
Joe Dreyer, BI Consultant at PBT Group
Reading a 2019 article by Kurt Cagle where he states “..graph databases have the potential to replace the existing relational market by 2030”, piqued my curiosity. On further exploration, the Gartner Hype Cycle for Artificial Intelligence, 2020 puts knowledge graphs at the “Peak of Inflated Expectations”. This means that organisations may be adopting this technology as ‘part of everyday work’ in the future.
So, what is a graph database and where does it fit into the organisation?
Before jumping into the technical side of things, it is important to understand knowledge and the management thereof. Of course, knowledge management is not new. One of the first results you get when googling knowledge is Ikujiro Nonaka. Nonaka proposed the SECI (Socialisation, Externalisation, Combination, and Internalisation) model as a knowledge conversion theory at organisations. You will also encounter words like Tacit- and Explicit knowledge, semantics, ontology, the list goes on.
Even though knowledge management ties into graph databases, graph databases do not necessarily mean knowledge management. Organisations need a knowledge management strategy and implementation plan to ultimately get value from the technology used. Furthermore, there needs to be an action on the knowledge used.
As an example, NASA adopted knowledge management as part of their way of work. By connecting people, tacit and explicit knowledge is shared and maintained, kept relevant, and actionable. NASA also uses Neo4J to manage knowledge for their human capital. By harnessing a graph database as part of the implementation, NASA links people, process, and a system to enable employees to be involved in future project opportunities. Ultimately, the employees and contractors contribute to this initiative because they want to. In this way, everybody wins in the organisation.
COVID-19 has changed the way we are working in such a way that knowledge management is now more important than ever. People sometimes feel alienated without the physical social interaction at work, companies struggle to share or obtain information. Add to the mix the real-time required knowledge and the complexity escalates.
When using applications like Modelangelo (a tool for the modelling and analysis of knowledge-intensive business processes) you will realise how complex a knowledge business process can be. And here knowledge graphs start to play a role.
This brings us back to the technical part where you will see companies like Microsoft, IBM, AWS, Google, SAP, Neo4j, Stardog, and Poolparty supply offerings for graph databases. When searching the use cases or benefits when using knowledge graphs, the list of companies and technologies keeps growing.
In summary, keep in mind that the technology mentioned in this article is used as the enablement for knowledge management. Knowledge management is the backbone from which the enablement evolves. The business requirement (value) needs to be there. So, do not let technology drive your knowledge management solution.
Start with the business side and decide what is the best out-of-the-box technical enabler. Consider open source as well. You should also do a small proof-of-concept to show the value. There are no silver bullets out there. Use the technology that is the easiest to use according to your needs.