Liar liar, pants on fire; or how to use subjective logic and argumentation to evaluate information from untrustworthy sources

摘要

This paper presents a non-prioritized belief change operator, designed specifically for incorporating new information from many heterogeneous sources in an uncertain environment. We take into account that sources may be untrustworthy and provide a principled method for dealing with the reception of contradictory information. We specify a novel Data-Oriented Belief Revision Operator, that uses a trust model, subjective logic, and a preference-based argumentation framework to evaluate novel information and change the agent’s belief set accordingly. We apply this belief change operator in a collaborative traffic scenario, where we show that (1) some form of trust-based non-prioritized belief change operator is necessary, and (2) in a direct comparison between our operator and a previous proposition, our operator performs at least as well in all scenarios, and significantly better in some.