Graph neural networks (GNNs) belong to the category of artificial intelligence, big data and smart data as well as industry and Factory 4.0. They are a special type of artificial neural network that is particularly good at handling information organised in the form of networks or relationships - so-called "graphs". Examples include social networks, supply chains or communication structures in companies.
Imagine you want to predict how information spreads in a large organisation. GNNs not only analyse individual employees, but also how they are connected and how they communicate with each other. This makes it possible to discover patterns in how knowledge is shared or how quickly news spreads.
A practical example from industry: a machine manufacturer wants to find out which spare parts in a factory network could fail first. With the help of graph neural networks, not only the individual machines are considered, but also their connections to each other in order to enable precise predictions and optimisations for maintenance and production.
Graph neural networks are therefore particularly valuable when it comes to analysing relationships between individual elements - a key technology for modern data analysis and networked systems.
Deutsch 
English (UK)