This is more like a controlled experiment: X received information L then 
decided to disconnect Y; L is the cause, disconnecting Y is the effect.

Now, let us assume that the network before L had weighted nodes and arcs. Can 
we explain why the effect was generated? What information can we infer about X?

Maybe X disconnected the link with Y because

  - X perceives Y as a threat
  - X does not trust Y
  - X  received information in the past from Y that does not wish to receive in 
the future
  - X did not really know Y and decided to maintain connections with 
people/groups he/she knows well
  - X was threaten by L and X fears caused the effect
  - X knows adversarial learning and decided to inject random noise by 
disconnecting with random links
  - X and Y agreed to deceive an observer by disconnecting online
  - X is crazy, walk up in the morning and decided to disconnect with Y
  - two or more of the above if the set does not include mutually exclusive 
reasons

Now, start from the initial belief (weights on nodes and links, and some meta 
data), write some simple rules/heuristics that map the meta data and weights to 
one or more of the above, run it!

The rules need to strengthen a hypothesis on the rational of the effect based 
on who else X disconnected with.

Bingo ... all those who initiated disconnected links (responded to L) revealed 
information that would have taken years of traditional social network analysis 
and dynamics to infer!

Reliability of L does not matter, what matters is that X responded :)

The real question now is, how to reconnect the network? Someone else needs to 
design this experiment :-)

Hussein

On 18/07/2013, at 8:28 AM, "Parks, Raymond" 
<[email protected]<mailto:[email protected]>> wrote:

In a strangely related story, Mark Zuckerberg reported that Facebook had the 
largest amount of unfriending and blocking ever on Sunday after the Zimmerman 
verdict.  ObComplexity: How does the dynamics of network connection breaking 
and making affect the analysis of networks?

Ray Parks
Consilient Heuristician/IDART Program Manager
V: 505-844-4024  M: 505-238-9359  P: 505-951-6084
NIPR: [email protected]<mailto:[email protected]>
SIPR: [email protected]<mailto:[email protected]> (send 
NIPR reminder)
JWICS: [email protected]<mailto:[email protected]> (send NIPR reminder)



On Jul 17, 2013, at 4:12 PM, [email protected]<mailto:[email protected]> 
wrote:

Glen writes:

"Lrnu, V oybpxrq rirelobql ba Snprobbx."

Here's some elisp code to keep them busy:

(defun annoy-nsa ()
 (interactive)
 (let ((start (point)))
   (spook)
   (rot13-region start (point))))



--------------------------------------------------------------------
myhosting.com<http://myhosting.com> - Premium Microsoft® Windows® and Linux web 
and application
hosting - http://link.myhosting.com/myhosting



============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com

============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
________________________________
THE UNIVERSITY OF NEW SOUTH WALES
UNSW CANBERRA AT THE AUSTRALIAN DEFENCE FORCE ACADEMY
PO Box 7916, CANBERRA BC 2610, Australia
Web: http://unsw.adfa.edu.au

CRICOS Provider no. 00100G

This message is intended for the addressee named and may contain confidential 
information. If you are not the intended recipient, please delete it and notify 
the sender. Views expressed in this message are those of the individual sender 
and are not necessarily the views of UNSW.
[http://www.openday.adfa.edu.au/images/openday_email_signature-2013.jpg]
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com

Reply via email to