Studies on Belief and Opinion Propagation in Multi-Agent Systems

2019 Fiscal Year Final Research Report

• Project/Area Number: 17K12746
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Intelligent informatics
• Research Institution: National Institute of Advanced Industrial Science and Technology
• Principal Investigator: Schwind Nicolas 国立研究開発法人産業技術総合研究所, 情報・人間工学領域, 研究員 (60646397)
• Project Period (FY): 2017-04-01 – 2020-03-31
• Keywords: Belief Change / Belief Revision Game / Brand Crisis Management / Belief Promotion / Public Announcement / Multi-Agent System / Propositional Logic / Network of Agents

Outline of Final Research Achievements

Daily, people share their opinions over social media. In particular, negative content regarding a brand could propagate rapidly and generate negative perceptions. So one needs to understand and monitor the public perception and the social reputation of companies. The "Belief Revision Games" (BRG) framework was introduced in 2015 for modeling such situations. This project aimed to enrich the BRG framework by providing a set of formal concepts to be integrated to it: (1) we explained what kind of revision policies the agents use, assuming that they revise their beliefs in a "rational" way; (2) we explained how an agent can be "manipulated", so as to convince all agents in the network of some "goal information" to be disseminated; (3) we introduced algorithms to simulate opinion propagation in a BRG.

Free Research Field

Knowledge Representation and Reasoning

Academic Significance and Societal Importance of the Research Achievements

本研究課題は国内外の研究者と共同で進め，人工知能分野の難関国際会議4件（3件はランクA*）と難関国際論文誌1件の発表を行った．