Date of Award
2010
Publication Type
Master Thesis
Degree Name
M.Sc.
Department
Computer Science
Keywords
Computer Science.
Supervisor
Kobti, Ziad (School of Computer Science)
Rights
info:eu-repo/semantics/openAccess
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.
Abstract
Population evolution algorithms such as Cultural Algorithms (CA) enable a global repository known as the belief space consisting of common cultural traits to influence the population space. Two important aspects of CA are the knowledge and its propagation. The population use social networks for communication. Knowledge representation is generally dependent on the application at hand. In this thesis the role of CA belief space knowledge in application neutral simulation is explored. A standard benchmark function is used to study the performance of evolutionary algorithms. The function captures the characteristics of a neutral world in dynamic settings. A multi-agent simulation was designed where autonomous agents are able to communicate, acquire and exploit various knowledge types including topographic, domain, historical and situational. While all these strategies showed improvements when searching for the global maximum, we found that domain based topographic exploitation strategies of the landscape were the more efficient.
Recommended Citation
Peiris, Viranthi, "Heuristics for Cultural Algorithm Knowledge Driven Search in Dynamic Social Systems" (2010). Electronic Theses and Dissertations. 334.
https://scholar.uwindsor.ca/etd/334