Work Product Descriptor (Artifact): Trust Model

This artifact describes the process by which trust values are derived from the experiences made by the agents.

Purpose

The trust model's purpose is to capture the semantics of the trust values and to describe the algorithm to derive them from the experiences.

Relationships

Input To	Mandatory: Design a Reputation System	Optional: Define Agent Decisions based on Trust Values Design Trust Measurement Infrastructure Identify Sources of Uncertainty	External: None
Output From	Define Trust Model Design a Reputation System

Main Description

It is helpful to use a formal notation to define the algorithm, as it avoids ambiguity and clarifies the semantics. A mathematical model can also easily be tested with simulated data to determine the accuracy of the model and to ensure correct behaviour in circumstances expected to arise in the system.

Properties

Optional
Planned

Tailoring

Impact of not having	Without this document, it will be difficult to comprehend the decisions made by the agents and to justify the semantics of the trust values.
Reasons for not needing	If a trust model from the literature is used without modifications, this document can be omitted if appropriate references are placed in other design documents.

More Information

Concepts	Trust in Multi-Agent Systems
Supporting Materials	Bibliography for Trust and Reputation in Multi-Agent Systems The Trust Life Cycle

This material is made available under the Creative Commons—Attribution-ShareAlike License v3.0.

© Copyright 2013, 2014 by Institute for Software & Systems Engineering, University of Augsburg
© Copyright 2015 by Chalmers | University of Gothenburg
Contributors
Contact: Jan-Philipp Steghöfer
Impressum

Parts of this material is based on the Eclipse Process Framework Practice Library, distributed under the Eclipse Public License V1.0.