I would like to compare my algorithm (the improved LPA) with Louvain, infomap and CNM (fast greedy) algorithms (available in Mat lab toolbox community) that has been implemented on LFR dataset.
I confronted with a problem when I cannot use the outputs of the algorithms for NMI criteria.
I will gratitude everyone could guide me about the matter!
Well first of all I have to ask you what do you use to test NMI.
A source code example implemented by Lancichinetti can be found at:
https://sites.google.com/site/andrealancichinetti/mutual
For this specific code, you need a specific output format of your results. In case you have a file as output, you can easily look through the small examples provided in the source code from the link above and see in what kind of format the file has to be. (each line represents the set of nodes from a community)
Fortunately, my problem was solved
But my question was, I couldn't use of the output of community detection algorithms that implemented in MATLAB , as the input at criteria NMI .
I work for this problem was solved
- Badges
- Science topic