Large networks (loosely known as graphs in mathematics) can be hard to visualise.
An intuitive method is to draw them as node-link diagrams. In a node-link diagram, the network entities are drawn as nodes, and the connections are drawn as lines. To draw these things, we need to find proper positions for the nodes. This is where graph layouts come into play.
A common way to lay out the nodes is to use a force-based simulation, and treat the nodes and links as a spring-mass system. While learning about this, I made a thing that demonstrates it (but is otherwise completely useless).