Relational Graph Neural Networks are at home in the fields of Artificial Intelligence, Big Data and Smart Data, and Digital Transformation. They help to recognise and understand very complex relationships in large data networks.
Imagine you have a vast network of data points – for instance, all the suppliers, customers, and products within a large company. Relational Graph Neural Networks enable computers to analyse the relationships between these points. They don't just think in boxes, but rather consider how everything is interconnected.
A simple example: A company wants to find out which customers have similar purchasing patterns. Relational graph neural networks take customer data as „nodes“ and shared purchases as „connections“ in what's called a graph. The network then learns to recognise patterns in these relationships – for instance, which customer might soon need a particular product.
This technique is used to gain entirely new insights from data, improve forecasts, and facilitate decisions – from industry to online shops.
Deutsch 
English (UK)