Formulating Manipulable Argumentation with Intra-/Inter-Agent Preferences

摘要: From marketing to politics, exploitation of incomplete information through selective communication arguments is ubiquitous. In this work, we focus on development an argumentation-theoretic model for manipulable multi-agent argumentation, where each agent may transmit deceptive others tactical motives. particular, study characterisation epistemic states, and their roles in deception/honesty detection (mis)trust-building. To end, propose the use intra-agent preferences handle inter-agent determine which agent(s) believe more. We show how argumentation agent, if detected, would alter agent's perceived trustworthiness, that affect judgement as should be acceptable.