Date of Award

2008

Publication Type

Master Thesis

Degree Name

M.Sc.

Department

Computer Science

Keywords

Engineering, Automotive.

Supervisor

Kent, Robert (School of Computer Science)

Rights

info:eu-repo/semantics/openAccess

Abstract

Cultural Algorithms (CA) offers a better way to simulate social and culture driven agents by introducing the notion of culture into the artificial population. When it comes to mimic intelligent social beings such as humans, the search for a better fit or global optima becomes multi dimensional because of the complexity produced by the relevant system parameters and intricate social behaviour. In this research an extended CA framework has been presented. The architecture provides extensions to the basic CA framework. The major extensions include the mechanism of influencing selected individuals into the population space by means of existing social network and consequently alter the cultural belief favourably. Another extension of the framework was done in the population space by introducing the concept of social network. The agents in the population are put into one (or more) network through which they can communicate and propagate knowledge. Identification and exploitation of such network is necessary sinceit may lead to a quicker shift of the cultural norm.

Share

COinS