Our force-directed layouts blog post looked ‘under the hood’ at the forces at work each time the powerful organic layout runs. In this post, we’ll look at each automatic graph layout that users rely on to detangle and make sense of their most complex networks.
Every one is available in our graph visualization toolkits.
This automatic graph layout is designed to display data that contains a clear sequence of distinct levels of nodes, or where information flows from one level to another.
Sequential layout examines link directions across the network and calculates which nodes belong to what level automatically. The biggest challenge is to minimize the number of crossed links between nodes at different levels. To do this, the algorithm places each node at the average position of its neighbors in adjacent levels, and then it swaps nodes in each level to reduce the number of link crossings.
If there are links between nodes in the same adjacent level, a third stage called ‘offsetting’ curves links above or below the nodes in between the connected nodes to avoid drawing behind them.
Once the nodes have been ordered in sequence, they’re distributed within each level to get the best-looking result for your data. You can choose to position them relative to each other based on connecting links, at regular intervals for a more symmetrical view, or stretched across the available width for networks with a lot of diagrammatic structure.
The stacking option makes it easy to visualize broader structures in sequential layouts. Stacking creates compact views of each tier at a much more readable zoom level, and saves space by pulling neighboring nodes into neat grids. Find out more about the stacking function
This is actually our third ‘force-directed’ layout.
Instead of running the simulation of the three forces (repulsion, springs and energy) straight off, it first bunches nodes together according to the structure of the network, i.e. nodes connected with the same set of nodes are grouped. Once the groups of nodes have been made, the force-directed algorithm runs, but operating on the groups instead of on individual nodes.
This positions each group of structurally similar nodes together, which reveals the graph’s structural composition. A great way to find node communities:
This automatic graph layout offers another way to structure layers of data. While sequential layout places nodes in distinct tiers, radial layout places layers in concentric rings, with the hierarchy’s ‘top’ node at the center.
This can be a great way to show dependency chains between generations of nodes. It’s particularly useful if your data contains many child nodes to each parent.
This layout also uses a circular pattern, but this time nodes are arranged according to connection density. Highly-connected nodes are positioned at the center of the network, and nodes with fewer connections are pushed to the edges.
The result is an attractive ‘fisheye’ lens effect that’s great for discovering the key nodes in large networks.
Visualize your own connected data!
If you have connected data and would like to visualize it for yourself, simply register for a free trial.
This post was originally published some time ago. It’s still popular, so we’ve updated the content to keep it useful and relevant.