Visit Homepage
Skip to content

Category: SNA Concepts & Metrics

Multiple edge types in Gephi

Ok, I know it’s summer, most of you are taking a well-deserved break, and there are a lot more fun things to do than dealing with networks. For those of you nodding fervently in agreement: shame on you! A plague on all your houses! Except Gryffindor, I’m in that one. And no, I didn’t just pick that one because Harry Potter‘s in it. I got sorted, the proper way. At You should really try it, it’s so much fun!   Anywho, I recently got a cry for help concerning Gephi, and I thought: there may be more lost souls out there, struggling to get this right. So here’s the pickle: can you visualize different types of edges in Gephi?   Not all your relationships are the same: you don’t hang with yo brudda of da same mudda in the same way as you do with your bros, and you don’t treat them the same as your hos. Like chicks before dicks, ya know? This was already true 100, 200, 500, even 1,000 years ago. People had family (through descent as well as marriage: sometimes it’s useful to distinguish between the two), friends, colleagues, superiors, inferiors, extraterrestrial acquaintances, … . Or you might want to look at different kinds of interactions between individuals: who writes/ lends money/ sells a slave/ … to whom, whatever, you name it! Although you’ll probably end up filtering your networks according to specific types of relations anyway, it’s not a bad idea to start out with the complete picture, to…

Continue reading Multiple edge types in Gephi


Zachary’s Karate Club Network

Anyone getting familiar with SNA will stumble upon Zachary’s karate club network sooner or later. Yes, it’s an actual karate club; no, it isn’t some exclusive SNA stress relief program and it isn’t, in fact, Zachary’s club. Zachary’s the guy who collected data on how the different members of a small university-based karate club interacted outside the club, so basically: did those dudes (dudettes?!) hang out during non-karate-related stuff? The Karate Club Network: visualisation of friendships The Karate Club Network: two groups He whipped up a nice network and as you can no doubt also see on the graph, he discerned two groups. One is centered around the instructor (#1 on the left), and one around the club president (#34 to the right). And, as in every group, there are some in the middle who just can’t decide. Or don’t want to, out of principle. Posers. But what’s interesting is that these factions were never recognized by the members themselves. As an experiment then, Zachary used Ford & Fulkerson’s maximum flow – minimum cut labeling procedure to determine to which group these crowd-pleasers actually leaned to. All hypothetically of course. The Karate Club Network: hypothesis turned reality Be careful what you wish for, they say. Well, in Zachary’s case, he may not have actually hoped for it, but it certainly didn’t harm his case: after a collision between the president and the instructor, the club split in, could you ever have guessed, two. The factions Zachary predicted in his model appeared…

Continue reading Zachary’s Karate Club Network

Leave a Comment

Community detection in networks and modularity

Today is Silke’s turn to shine again, this time with an intelligent bit on community detection! I have just one thing to say: Good moaning guv! DataNinjas are back, bringing the best of SNA to you all the way from Fog City aka London! While Yanne withers away back home in Leuven, I have been trying very hard to drink my sorrow away in the many pubs with my new (and I must admit: v. amusing – just in case they’re reading this *wink wink*) colleagues.   Cheers, sweetie darling!   Community Detection For the few hours that I’ve been sober though, I’ve been focusing on another aspect of SNA: community detection. Not meaning that I’ve been walking around in the city with my binoculars, spying on the artistic, academic, queer communities – although I have, obviously. We’re – of course – talking about community detection in a network, or by that we mean: taking a look at the structure of the network by focusing on its underlying sub-units, which are made up of highly interconnected nodes. In Gephi, you can automatically get the software to divide your network in different communities. It also provides you with a value for modularity. This value (between -1 & 1) measures the density of the edges inside the communities in comparison to the density of the edges between the different communities. So it’s actually a measure for how good (or bad) the chosen community detection method is. These features are really nice, colouring the…

Continue reading Community detection in networks and modularity

Leave a Comment