I suggest you use symbolic IDs, i.e. vertex names instead of the internal
numeric ids.

GAbor

On Fri, Mar 13, 2015 at 12:42 PM, Daniele Rotolo <[email protected]>
wrote:

>  Hello,
>
>   I have publication data as follows:
>
> *Organization_name Publication_id Publication_Year*
>
> Organization 1 001 2013
>
> Organization 2 001 2013
>
> Organization 1 002 2012
>
> Organization 3 002 2012
>
> ...
>
>
>  I can easily convert this 2-mode network (Organization-Publication) in a
> 1-mode network (Organization-Organization) by using R and then graph. I
> managed to do this also over time, by filtering the 2-mode network by date.
> However, this approach inevitably assignes different IDs to the same
> organization over time. I would like instead to produce networks that have
> a constant number of nodes (i.e. nodes' IDs remain the same across time)
> where nodes can leave and join the network as well as edges can change over
> time.
>
> Any suggestion on how to do structure the data in this way?
>
> Many thanks!
> Daniele
>
> _______________________________________________
> igraph-help mailing list
> [email protected]
> https://lists.nongnu.org/mailman/listinfo/igraph-help
>
>
_______________________________________________
igraph-help mailing list
[email protected]
https://lists.nongnu.org/mailman/listinfo/igraph-help

Reply via email to