Engineering Social Agent Creation into an Opportunity for Interviewing and Interpersonal Skills Training: Socially Interactive Agents Track
Engineering Social Agent Creation into an Opportunity for Interviewing and Interpersonal Skills Training: Socially Interactive Agents Track
复制标题
将社交代理创建转化为面试和人际技能培训的机会：社交互动代理轨道
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
Benjamin C. Lok
中科院分区：
文献类型：
--
作者：
Shivashankar Halan;I. Sia;A. Miles;M. Crary;Benjamin C. Lok
使用智能的，互动的社会代理人的临床面试和人际交往技能的培训，在医疗保健教育已被观察到正在增加。然而，能够快速和可扩展地创建强大而多样化的智能社会代理，这些智能社会代理可以出于教学原因整合到教育课程中，这仍然是一个挑战。在本文中，我们提出了一种新的方法来创建虚拟病人（社会代理人，发挥病人的作用），通过重用以前的学生创建的互动社会代理人的对话语料库信息。在这种方法中，作为人际交往技能培训练习的一部分，医疗保健学生创建自己的虚拟患者。这些虚拟病人代理人创建被证明是有效的工具，以培训其他学生在未来与他们的面试和人际交往能力。通过在六年多的时间里将虚拟患者创建练习整合到七门卫生专业课程中，我们证明了医疗保健专业的学生可以创建强大而多样化的虚拟患者社交代理，这些代理可以用作教学工具，并且在创建过程中还可以提高他们自己的临床面试和人际交往技能。
The use of intelligent, interactive social agents for clinical interviewing and interpersonal skills training in healthcare education has been observed to be on the increase. However, enabling rapid and scalable creation of robust and diverse intelligent social agents that can be integrated into educational curriculum for pedagogical reasons is still a challenge. In this paper, we present a novel approach for creating virtual patients (social agents that play the role of a patient) by reusing conversational corpus information from previous student-created interactive social agents. In this approach, healthcare students as part of an interpersonal skills training exercise create their own virtual patients. These virtual patient agents created are demonstrated to be effective tools to train other students in the future with their interviewing and interpersonal skills. By integrating virtual patient creation exercises in seven health professions courses over six years, we have demonstrated that healthcare students can create robust and diverse virtual patient social agents that can be used as pedagogical tools and in the process of creation also improve their own clinical interviewing and interpersonal skills.