Artificial Intelligence Trust Modeling in Multiagent Systems to Streamline Social Networking

多代理系统中的人工智能信任建模可简化社交网络

• 批准号: RGPIN-2016-03615
• 负责人: Cohen, Robin
• 金额: $ 2.77万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2016
• 资助国家: 加拿大
• 起止时间: 2016-01-01 至 2017-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=614497
• 关键词: Artificial; Intelligence; Trust; Modeling; Multiagent

In this research, we aim to address the current topical but critical problem of information overload arising from the use of computers, especially in social networking environments. Our approach is to leverage artificial intelligence techniques: to develop novel paradigms for trust modeling, user modeling and multiagent cooperative coordination which will provide longstanding advances to theoretical research while at the same time forming the basis for significant breakthroughs in the management of our online existence. A companion aim is to embrace the current trend in artificial intelligence to leverage applied research towards the most valuable new insights in the development of theories and models. Towards this end, we will introduce several key application areas (healthcare, education and transportation) as motivators and as testbed areas for validation of our approach.Central to the research will be novel methods for trust modeling in multiagent systems, including user-specific trust evaluation (specific to the trustor), algorithms for trustees to engender trust, reasoning about when to solicit trust values from peers (the value of coordinated waiting), and projecting trust modeling into a specification of agent decision making.In all, we will not only advance artificial intelligence research but will also offer to Canada techniques of critical use to all organizations and individuals to enable more effective online communication and coordination between users. This in turn will translate into significant savings in time and in stress, towards an improvement of our overall state of wellbeing, to improve our overall economy. Relevant everyday contexts include massively open online courses, electronic marketplaces, online environments for social communication and peer-based networking to enable self-help. We are also interested in enabling enhanced access to online social networks for users with assistive needs (including the elderly).

在这项研究中，我们的目标是解决当前热门但关键的问题，即信息过载，这是由计算机的使用引起的，特别是在社交网络环境中。我们的方法是利用人工智能技术：开发信任建模、用户建模和多智能体协作协调的新范式，这将为理论研究提供长期的进展，同时为我们在线存在管理的重大突破奠定基础。另一个目标是拥抱人工智能的当前趋势，利用应用研究在理论和模型的发展中获得最有价值的新见解。为此，我们将引入几个关键应用领域（医疗保健、教育和交通），作为验证我们方法的激励因素和试验台领域。该研究的核心将是多智能体系统中信任建模的新方法，包括特定于用户的信任评估（特定于委托人）、受托人产生信任的算法、何时从同伴处请求信任值的推理（协调等待的价值），以及将信任建模投影到代理决策规范中。总之，我们不仅将推进人工智能研究，还将向加拿大提供所有组织和个人使用的关键技术，以实现用户之间更有效的在线沟通和协调。反过来，这将转化为时间和压力的显著节省，从而改善我们的整体健康状况，改善我们的整体经济。相关的日常环境包括大规模开放的在线课程、电子市场、用于社交交流的在线环境以及能够实现自助的基于同伴的网络。我们也有兴趣为有辅助需求的用户（包括老年人）提供更多访问在线社交网络的机会。