Analyzing Socio-Economic Complex Adaptive Networks: A Hybrid Approach

Abstract

In this thesis, we aim to explain the structural changes within socio-economic complex adaptive networks with respect to social characteristics of individuals. Actors of socio-economic complex adaptive networks are social units that undergo various strategic processes to achieve their goals. So there is a need to study such networks systematically with respect to their actors and their strategic interactions as it evolves over time. Variety of stochastic and strategic network formation models has been introduced in the literature to explain the emergence of certain networks characteristics. However, each of those techniques has their own limitations. While both stochastic and strategic network formation models are able to tell us what the interesting characteristics of the network are, the extent to which those characteristics affect the outcome of individuals and the whole society is still unclear and has received little attention in the literature. Our focus is on human-to-human communication environments, where the process of link establishment among network members is not random and the process of a network growth requires a proper economic inventive modeling among its members. We aim to show that a proper incentive modeling is needed for the process of link establishment within socio-economic complex adaptive networks and its effects will be reflected on the emerging networkcharacteristics as well as the networking outcome (i.e., learning outcome or utility gain at the individual level). Networking outcome can be considered as whatever an individual gains out of his or her connectivity within a network (i.e., learning outcome or utility gain at the individual level). That is to say, individual actions determine the network structure and similarly structure also influences individual actions and thinking. They constrain and enable actions. Therefore, there is a feedback loop between individual actions and network structure. We argue in this thesis that, a hybrid approach based on complex adaptive system theory is needed for studying socio-economic complex adaptive networks. Consequently, we can justify why the underlying network structure is constantly changing and as the result a certain type of network with specific characteristics emerges. The structural changes (emerging network characteristics) are the changes in the clustering coefficient value and the average shortest-path length. For capturing, comparing, and explaining the structural changes and outcome of individuals within socio-economic complex adaptive networks, we developed a multi-agent based model. With the help of agent-based modeling, we are able to test and evaluate this approach.