(Towards) Finding Suitable Analysis Techniques for Agent-based Networks
Generated from Social Processes

By: Shah Jamal Alam and Ruth Meyer
Date: 2nd August 2006
CPM Report No.: CPM-06-164

Presented at the Social Network Analysis: Second Forum on Advances and Empirical Applications, Leeds, June 30-July 1 2006.


HIV/AIDS has long been understood as the major obstacle in human development, especially in the poorest regions such as the sub-Sahara countries in Africa. However, most people in this region regard other stressors, e.g. food insecurity, high increase in adult deaths, unavailability of irrigation water, lack of grants, etc. as the major causes of their distress. This is part of an ongoing project that attempts to model the complexity that arises as a result of actors’ interactions at the individual and the household level. The model addresses several issues in understanding how such stressors, in particular, HIV/AIDS, affect the community structure. Networks that result from social processes in simulation are dynamic and co-evolving in nature. In this paper, we outline some of the important issues that arise in dynamic social networks which result from social processes and other exogenous factors. We suggest a scheme as a way forward in understanding and analyzing such networks that change over time.

 Keywords: agent-based social networks, dynamic social networks, evidence-based modeling

Accessible as: