Communication and interaction design for chronic health situations

对话系统研究的日益成功使得体现会话代理成为一个完美的候选人标准，在人机交互，包括健康相关的辅导和援助ingroupconversation。交际行为的自然性为用户提供了一个舒适的互动环境。在对话系统中使用多个通信通道来模拟人工智能体中的人性已经取得了许多进展。本研究计划将研究人类在与这些辅助技术进行自然交互时的行为，以创建一个智能会话Agent（ECA），使用户能够通过简单地与ECA对话并从ECA接收答案来有效地实现他们的目标。我们的目标是建立一个智能虚拟Agent（IVA），它在正确的时间自动介入，并根据用户的条件和兴趣模型向用户提供内容和信息。拟议的研究将开发新的理论、模型、算法和技术，使用户更容易以更少的认知努力更自然地进行交互。这项工作也将提高对这些系统如何影响人类的基本理解。高素质人员（HQP）将接受人工智能（AI）和人机交互（HCI）的培训，特别是非自然交互模型。在使用社交媒体组来帮助有相关健康问题的有想法的用户的社交健康平台中，一个很大的挑战是有许多健康状况，因此手动为这些状况中的每一个建立和开发社区和服务是不可行的。解决方案将是开发一个AI个性化平台，提供可扩展的，可持续的和自动化的服务和策略。因此，本文的研究工作是建立一个个性化的智能虚拟伴侣（Agent）的原型，它将通过对话收集关于用户的信息，建立用户模型，并将通过对话以特定的和定制的方式与用户进行交互，提供个性化的指导，培训和帮助。

The growing success of dialogue systems research makes embodied conversational agents a perfect candidatefor becoming a standard in human computer interaction including for health related coaching and assistance ingroup conversation. The naturalness of communicative acts allows for providing a comfortable ground for theusers to interact with. There have been many advances on using multiple communication channels in dialoguesystems in the way of simulating humaneness in an artificial agent. The proposed research program will studyhuman behavior while interacting with these assistive technologies using natural interaction methods in orderto create an Embodied Conversational Agent (ECA) that enables users to efficiently achieve their goals throughsimply talking with the ECA and receiving answers from it. We aim to build an Intelligent Virtual Agent(IVA), which automatically intervenes at the right time and provides content and information to the user basedon user's condition and interest models. The proposed research will develop new theories, models, algorithms,and technologies to make it easier for users to interact more naturally with less cognitive effort. This work willalso improve the fundamental understanding of how such systems affect humans. Highly qualified personnel(HQP) will be trained in artificial intelligence (AI) and human computer interaction (HCI), particularly innatural interaction models. Within social health platforms using say social media groups to help like mindedusers with a related health problem, one big challenge is there are many health conditions, so it is not feasibleto build and develop community and services for every one of these conditions manually. The solution wouldbe to develop an AI personalized platform which provide scalable, sustainable and automated services andstrategies. Therefore, this research effort is to build a prototype of a personalized intelligent virtual companion(Agent) that will collect information about the user via conversations, build a user model and will interact withthe user in a specific and customized way via conversation which provides personalized coaching, training andassistance.