| Abstract |
Models that capture salient features of the structure and the nature of competition in differentiated web markets are proposed in this thesis. The main objective is to examine evolutionary dynamics in the formation of web geography and economy by studying an agent-based experimental environment. Theoretically, in a world with no transportation costs, such as the virtual world, social and economic activity should be more or less evenly distributed across the Internet landscape. In reality, however, the web economy illustrates a dramatic unevenness. Internet users, while able to live an exciting online experience from the permanent exploration of a really immense web, they like to aggregate in particular places which gradually became extremely popular. To explain such a spectacular concentration of web habitants in a handful of sites, we have to accept that, exactly as in the real world [1], these virtual concentrations are not only the result of inherent differences among locations, but, also, the result of some set of cumulative forces, necessarily involving increasing returns and various agglomeration economies. In the complex and constantly evolving environment of the Web, fierce competition for visitors/consumers arises among sites. The number of sites increases with exponential rate and so does the number of web users. A wide variety of new products and services is made available in an attempt to capture the users attention. Web sites have to engage into a strategic behavior in order to survive and prosper. Two agent-based economic behavior models are proposed in this thesis, which capture the most essential characteristics of the Web economy. Heterogeneous and autonomous agents that mimic the behavior of web sites and users were designed based on empirical observation, statistics and case studies. The models were implemented in an agent-based environment on top of the Mozart platform. This environment enables us to study the dynamics of local interactions and individual behaviors which produce non-predictable global structures through time. The first model is a general model for studying the structure and competition that emerges in a virtual economy comprised of many differentiated markets. The second model, which is yet in an early stage of development, focuses more on the role of the Internet as a distribution channel. B-2-C sites compete in a differentiated market for consumers and form coalition in order to increase profits. Results from various experiments based on these models are presented in order to broaden our understanding of the mechanisms that create the structure of web markets and contribute to the survival and success of web sites.
|
Βασισμένα σε πράκτορες μοντέλα αγορών διαφοροποιημένων αγαθών του Διαδικτύου
