SocioScape - Spatio-temporal Visual Analysis of Group Dynamics in Social Networks


Authors: Reda, M. K.

Publication: Submitted as partial fulfillment of the requirements for the degree of Master of Computer Science, University of Illinois at Chicago, Graduate College

URL: http://www.evl.uic.edu/kreda/Thesis_KhairiReda.pdf

This thesis presents SocioScape, an interactive visualization tool that embodies a methodology for the visual analysis of spatial and temporal group dynamics in social networks. The methodology introduces a novel visual representation technique suitable for dynamic social networks. This representation provide an advantage over dynamic graphs by explicitly illustrating the evolution of social groups and association choices made by actors over time. The representation is combined with a well-established technique for depicting spatio-temporal data, allowing analysts to investigate the effect of the physical positioning of actors and their movement in the environment on their social behavior. This integration also facilitates the investigation of potential hypotheses that explain the emergence of the observed social structure structure. A case study demonstrates the usefulness of the tool.

The primary contributions of this thesis include:
  1. A novel visual representation method for dynamic social networks that departs from traditional graph-based visualizations, revealing the evolution of social groups and association choices that actors make over time.
  2. A methodology that integrates abstract representation of social interactions with a spatio-temporal visualization to allow analysis of the role of environment in shaping the underlying social structure.
  3. A case study in which the methodology was used by behavioral ecologists to explore the social behavior of two animal populations of endangered species.


Date: July 16, 2009

Document: View PDF
Screenshot of SocioScape showing the movement and social structure of Grevy’s zebra herds - K. Reda, EVL

Related Entries

Directory:

Events:

Research:

Related Categories