Print

Print


*****  To join INSNA, visit http://www.insna.org  *****

Not exactly sure what you mean by this, but there are many methods of over
time analysis. The biggest issue is the data and the research question. If
you are interested in identifying behaviors that drive network evolution,
you can use Siena and there is a new time test for heterogeneity (Lospinoso
et al, 2010).  
 
Siena is agency-oriented while the evolution of networks takes place at the
systems level. The networked system evolves in terms of a large possible
sets of interactions and recursions. (Anticipations are additionally
possible.)
 
At each moment of time one has a network which can also be written as a
matrix. Over time, one thus obtains a set of stracked matrices. The formal
description of the evolution in the network would assume solving a set of
partial differential equations. The problem is usually not analytically
solvable, and the issue is thus how to approach it. 
 
In the dynamic version of visone, time is used as another variant and the
stress-minimalization is extended from two to three dimensions. Instead of
reducing (Kruskall's) stress at each moment of time, one can also reduce
stress in relation to preceding or subsequent years. When more years are
included, the animation becomes more stable (Baur & Schank, 2008;
Leydesdorff & Schank, 2008). This solution has become possible because of
optimizing the majorant (Erten et al., 2004; Gansner et al., 2005).

I understood that there is also a solution to the problem in ORA. Pajek2SVG
uses interpolation between the years. In Sonia, new nodes are not positioned
in anticipation to the next year.
 
Best wishes, 
Loet
 
References:
Baur, M. & Schank, T., 2008. Dynamic Graph Drawing in Visone. Technical
University Karlsruhe, Karlsruhe. Available at
<http://digbib.ubka.uni-karlsruhe.de/volltexte/1000007990>
http://digbib.ubka.uni-karlsruhe.de/volltexte/1000007990 (Retrieved on  June
18,  2010).

Erten, C., Harding, Ph. J.,  Kobourov, S. G., Wampler, K., & Yee, G. V.
(2004). GraphAEL: Graph animations with evolving layouts. Pages 98-110 in
Liotta, G., Editor, Graph Drawing, Perugia, Italy, September 21-24, 2003;
Springer, 2004. 

Gansner, E. R., Koren, Y., & North, S. (2005). Graph Drawing by Stress
Majorization. In J. Pach (Ed.), Graph Drawing, Lecture Notes in Computer
Science (Vol. 3383, pp. 239-250). Berlin/Heidelberg: Springer.

Leydesdorff, L., & Schank, T. 2008. Dynamic Animations of Journal Maps:
Indicators of Structural Change and Interdisciplinary Developments. Journal
of the American Society for Information Science and Technology, 59(11),
1810-1818.

 

 

 If you want to compare two networks in time to see if they are different,
you can use an ERGM with edgecov term for the earlier network. If you have a
lot more data as in email networks, you can monitor/plot various network
statistics over time, or to be more sophisticated, you can assume a range of
typical behavior and use SPC methods to detect departures from this over
time (McCulloh, 2009). Again with enough data, Fourier Analysis can be used
to identify significant periodic trends in network statistics (McCulloh,
Johnson, Carley, n.d.) which can be used to fine tune assumptions in SPC
approaches. SPC approaches can also be useful for exploring area of time
heterogeneity when Siena models won't converge. Jaccard Index can also be
used to determine network change along with Hamming and Euclidean distance.
Another approach that is somewhat controversial is to compare triad census
of networks, but suppose you could really compare many different network
statistics over time. This is why the research question is so important.

Hope this helps.

Ian 

Sent via BlackBerry by AT&T

  _____  

From: "Loet Leydesdorff" <[log in to unmask]> 
Date: Thu, 17 Jun 2010 13:50:20 +0200
To: 'Ian McCulloh'<[log in to unmask]>; <[log in to unmask]>
Subject: RE: How to visualize the evolution of social network?


Yes, the issue is whether there is a statistics for controling change as an
alternative to comparative statics (with interpolation).
 
Best, Loet
 
  _____  


Loet Leydesdorff 
Amsterdam School of Communications Research (ASCoR), 
Kloveniersburgwal 48, 1012 CX Amsterdam. 
Tel.: +31-20- 525 6598; fax: +31-20- 525 3681 
 <mailto:[log in to unmask]> [log in to unmask] ;
<http://www.leydesdorff.net/> http://www.leydesdorff.net/ 

 

 


  _____  

From: Social Networks Discussion Forum [mailto:[log in to unmask]] On
Behalf Of Ian McCulloh
Sent: Thursday, June 17, 2010 1:14 PM
To: [log in to unmask]
Subject: Re: How to visualize the evolution of social network?


***** To join INSNA, visit http://www.insna.org ***** ORA has network
visualizations over time.  It also has features such as statistical process
control to detect changes over time, Fourier analysis to identify
periodicity in over time data.  ORA is freely available at
www.casos.cmu.edu.


2010/6/17  <[log in to unmask]>


***** To join INSNA, visit http://www.insna.org ***** I have a small online
social network embedded with the time variable on the second scale. 
Does anyone know any software dealing with the visualization of the network
evolution?  
I 've tried Pajek and UCINET, both of which could only draw the network
graph temporally.

Thanks for your generous help. 

ZHANG Lun
PhD Candidate,
City University of Hong Kong  
_____________________________________________________________________ SOCNET
is a service of INSNA, the professional association for social network
researchers (http://www.insna.org). To unsubscribe, send an email message to
[log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of
the message. 


_____________________________________________________________________ SOCNET
is a service of INSNA, the professional association for social network
researchers (http://www.insna.org). To unsubscribe, send an email message to
[log in to unmask] containing the line UNSUBSCRIBE SOCNET in the body of
the message. 

_____________________________________________________________________
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.