Date of Award:

5-2014

Document Type:

Thesis

Degree Name:

Master of Science (MS)

Department:

Computer Science

Committee Chair(s)

Vicki H. Allan

Committee

Vicki H. Allan

Committee

Nicholas Flann

Committee

Dan Watson

Abstract

Interaction between agents is one of the key factors in multi-agent societies in order to cooperatively execute complex tasks which are beyond the capability of a single agent. In the self-adaptation society, agents try to keep best neighbors around themselves. Agents use history of past iterations to evaluate their neighbors and locally modify their structural links. Main characteristics of this model are its decentralization, dynamicity and no need of external control. This research at first deals with implementing self-adaptive agent organization. Then it focuses on evaluating cooperation peers in decision making. Trust as an evaluation mechanism lies at the core of all interactions in the agent organizations. Agents build trust about a target based on their direct experiences and third party recommendations. Since agents have different perception of trust, context of decision should be taken into account. Uncertainty is one of the main contextual factors that affect agents’ decisions. The aim of the second work is helping agents have better evaluation in decision making utilizing recommendation-based trust and adaptive risk.

Checksum

0c94a6a0cd9dfa5e9cba8058f0a22f30

Share

COinS