Return to search

Modelování rozhodovacích sítí / Modelling of decision-making networks

Delegative democracy is a new concept of democratic governance. It is meant as a new system of trust among people. It is derived from principles of direct democracy and presents only minimal form of representation. The voter is not forced to cast his vote directly, but it can be transferred to a delegate. Delegative democracy is based on openness, awareness and trust. It is still only a concept and for real application it has to face rising critics. Main issues are high costs, threat of populism and corruption and overall complexity. Agent-based modelling is chosen to test the validity of delelagative democracy principles. It is suitable for studying social phenomena such as democratic governance. Agent-based models apply a "new kind" of artificial intelligence -- a dynamic system composed of individual and autonomous units -- agents -- which interact within the environment. Universal methodology called Agentology is used to propose a new agent-based model. This methodology is composed of subsequent steps. These steps cover initial analysis, conceptual and technological proposal and development of the model itself. Assessment of delegative democracy principles is achieved with data acquired from the agent-based model. Delegative democracy is less effective for small and cooperative system than direct democracy. If more voters are delegating, this effectiveness decreases even more. This conclusion is based on initial parameters of the model. Delegating voters count is indeed a relevant parameter whereas total voters count is not. The model has its basic predicative value and is open to further elaboration.

Identiferoai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:124786
Date January 2011
CreatorsŠilar, Pavel
ContributorsBurian, Jan, Šalamon, Tomáš
PublisherVysoká škola ekonomická v Praze
Source SetsCzech ETDs
LanguageCzech
Detected LanguageEnglish
Typeinfo:eu-repo/semantics/masterThesis
Rightsinfo:eu-repo/semantics/restrictedAccess

Page generated in 0.011 seconds