An Operational Architecture for Knowledge Graph-Based Systems
Matthias Sesboüé, Nicolas Delestre, Jean-Philippe Kotowicz, Ali Khudiyev, Cécilia Zanni-Merk · Procedia Computer Science · 2022
Knowledge Graphs (KG) are gaining in popularity recently, notably since big tech giants announced they are using the technology. While the term is becoming popular, it is not new, and its ideas are even older. The research community has extensively studied knowledge Graphs in their various forms. Furthermore, the approach has been applied and proved valuable in many different applications. However, we found a lack of papers presenting the integration of KGs in a system regardless of the downstream application. We explore how KGs can fit in an overall information system independently from any specific use case, i.e., what we will consider knowledge consumption. We propose an architecture to understand better the KG roles within a system and how they can be integrated and implemented in a business context. We introduce each element of the latter architecture and discuss some candidate technology to implement them. Our work implements Knowledge Graph-Based Systems considering the constraints of a small to medium-sized enterprise.