Fixed nodes in visualization of multiplex networks

[genja fu]

Hi

Doing a multiplex network analysis I wanted to visualize 6 networks (3 types of relations of the same actors / nodes at 2 points of measurement). However, I could not find a way to position the nodes at the same places in all 6 visualizations (locked positions) in a way that the networks can easily be compared. Is there any way to do this in visone (I'd prefer the circular layout)? I would be very grateful if you could help me with this issue.

[Felix Schönenberger]

As far as I know, all the layouts do optimize for something. The position
will not only depend on the node index but on the structure (minimize stress,
edge crossings, ...). Since the structure changes the node position changes.

To really fix the layout you could attach x and y attributes to your data and
map those attributes to the coordinates. Creating the coordinates of the
circular layout should be easy in any scripting language. Or you can use one
network as a template: Layout, use
`visualization > appearance to attribute > coordinates`, copy the attributes
to all networks, apply via `mapping > coordinates > cartesian`.

On 19-06-02 15:18, gen...@gmail.com wrote with possible deletions:

[genja fu]

Hi Felix,

Thank you very much for reply. Unfortunately, I wasn't able to "copy the attributes to all networks"? How would I do it?

Anyways, I was able to produce 6 networks with the same node positions by creating a network collection and producing a dynamic layout as described here:

https://visone.info/wiki/index.php/Collections_(tutorial)#Dynamic_visualization_and_animation_of_a_network_collection

By using the anchored method (and a reference layout) and putting stability to 100 % I was able to get the visualisations I was seeking.