Wednesday, April 18, 2012

Special session: Computational Intelligence and Social Media

The below is a call for papers for a special session of the 2012  IEEE Workshop on Computational Intelligence for Security and Defence Applications (IEEE CISDA) 2012. The deadline for submitting papers for this special session is 23 April, 2012. This conference will be held in Ottawa, Canada, 11-13 July 2012.


Computational Intelligence and Social Media

Organizers: John Verdon, DRDC Ottawa, Canada
Contact: john.verdon -at- drdc-rddc.gc.ca

Military and security communities are hard-pressed to develop the capabilities required to exploit the huge volumes of data, the new forms of information, and rapidly changing content of Social Media (SM) such as blogs, wikis, videos, social graph based systems (such as Facebook, Twitter) and many other SM systems that are being deployed. SM is also in its infancy, so there is a huge potential for SM to evolve far beyond its current capabilities and types of information.

Computational Intelligence techniques (Neural network, Evolutionary computation, Fuzzy Systems, Particle Swarms, etc) have often been based on, and and have been related to, highly complex, structured, and dynamic natural systems in [biology, neuroscience, brain, psychology, sociology]. This may make them particularly well suited for the extraction of intelligence from existing forms of SM, for the [modeling, prediction, control] of SM activity, as well as for providing some capability of keeping up with rapidly evolving and new forms of SM. Papers that deal with massive datasets are of particular interest, and, naturally, papers should relate to defense and security needs, applications, and tool-sets. Security & Defense needs and Social Media [some examples from Forrester 2011, Verdon 2012] topics include, but are not limited to:
  • Language translation - filtering a collection of documents down to those that should be translated by humans
  • Knowledge extraction - validating facts from unstructured and questionable sources
  • Document summarization - extracting the sense of a document or a group of topically-related documents, and establishing the main points of consensus and divergence
  • Trend identification - as well as possible causal linkages among trends and supporting evidence
  • Active learning - determining where information is lacking and which data would be most productive to acquire
  • Security - at what threshold does secrecy become a liability rather than an asset for security?
  • Reputation and/or Recommendation - ‘quick trust’ of the participants and possibly of the information