File manipulation, stats, metrics - various questions
Posted: 23 Nov 2017 17:17
Hi guys! I have three questions i would like to ask!
1.) what is the best way to combine several networks, each contained within one .gephi file?
2.) how can i extract (or make list), say for the top 5-10 nodes for each calculated centrality metric?
3.) can someone explain difference between eigenvector and page rank in the context of Gephi because these metrics are very similar.
Additional explanation regarding #1:
I have 1st degree connections for over 100 Facebook pages and those pages are separated in 10+ categories regarding their type/content.
For each category I've made one .gephi file after I've imported certain number of .graphml files (one page has it's own .graphml) to make one category.
The thing is that I want to analyse each category separately and than everything together.
For some reason if I make one category from several .graphml files nodes/edges are not overwritten (except if they are the same), they are getting unique ID each.
On the other hand if I want to group all made categories from several .gephi files those nodes/edges already have their ID's and those ID's are overwritten.
So the question is if I can get the complete network without combining 100+ .graphml files all over again and use .gephi files instead, BUT to avoid ID's being overwritten...?
Thanks!
1.) what is the best way to combine several networks, each contained within one .gephi file?
2.) how can i extract (or make list), say for the top 5-10 nodes for each calculated centrality metric?
3.) can someone explain difference between eigenvector and page rank in the context of Gephi because these metrics are very similar.
Additional explanation regarding #1:
I have 1st degree connections for over 100 Facebook pages and those pages are separated in 10+ categories regarding their type/content.
For each category I've made one .gephi file after I've imported certain number of .graphml files (one page has it's own .graphml) to make one category.
The thing is that I want to analyse each category separately and than everything together.
For some reason if I make one category from several .graphml files nodes/edges are not overwritten (except if they are the same), they are getting unique ID each.
On the other hand if I want to group all made categories from several .gephi files those nodes/edges already have their ID's and those ID's are overwritten.
So the question is if I can get the complete network without combining 100+ .graphml files all over again and use .gephi files instead, BUT to avoid ID's being overwritten...?
Thanks!