The term "explainable graph neural networks" originates from the fields of artificial intelligence, Industry and Factory 4.0 and Big Data and Smart Data.
Graph neural networks are a special form of artificial intelligence that is particularly good at handling data that can be mapped in networks or relationships - for example, the interaction of different machines in a factory or the connection of customers in a social network.
It is often difficult to understand why such AI systems make a certain decision. This makes it risky for companies to trust these systems. "Explainable graph neural networks" means that these AI models are structured in such a way that humans can understand their decisions.
A simple example: Imagine an AI monitors all the machines in your production line and issues a warning before a machine breaks down. An explainable system can show you exactly why it thinks this machine will have a problem - for example, because similar problems have previously occurred with machines with the same characteristics. This allows you to better assess how reliably the AI is working and how you should react.
Deutsch 
English (UK)