A Computational Model of Trust Based on Message Content and Source
Abstract
We propose a general possibilistic framework to determine the a-gent's trust degree in a source, starting from the content of the messages such source provides and based on the beliefs of the agent about the capability of the source to provide " useful information ". The result is a framework with unique characteristics, which combines experience-, reputation-, content-, and category-based models of trust in one coherent computational model.
Domains
Artificial Intelligence [cs.AI]
Origin : Publisher files allowed on an open archive
Loading...