客户服务中人工智能情感胜任力模型和测量指标构建  

The Emotional Competency Model and Measurement Index for Artifi cial Intelligence in Customer Service

作　　者：邓士昌 田芊[2] 林晓祥 DENG Shi-chang;TIAN Qian;LIN Xiao-xiang(School of Management,Shanghai University of International Business and Economics,Shanghai 201620,China;Fudan University,Shanghai 200433,China;School of International and Public Aff airs,Shanghai Jiaotong University,Shanghai 200240,China)

机构地区：[1]上海对外经贸大学工商管理学院,上海201620 [2]复旦大学社会发展与公共政策学院,上海200433 [3]上海交通大学国际与公共事务学院,上海200240

出　　处：《上海对外经贸大学学报》2020年第4期100-110,共11页Journal of Shanghai University of International Business and Economics

基　　金：教育部人文社会科学基金项目“认知偏差因素对风险性网络借贷消费行为的影响机理及干预机制研究”(项目编号:19YJC630027);上海市哲学社会科学规划课题“非理性互联网金融投资行为中的认知偏差因素:作用机理及干预机制研究”(项目编号:2017EGL006);上海市晨光计划项目“网络偏差信息行为的心理机制及治理研究”(项目编号:16CG61)资助。

摘　　要：人工智能正在快速应用于客户服务领域,然而多项研究却指出用户对AI服务的体验并不满意。人们普遍认为AI客服不如人类服务者"善解人意",但对于服务型人工智能应该拥有什么样的"情商"能力,现有研究却缺乏探讨。本文采用"他评"的思路,基于757名用户调查数据结果,构建了拟合良好的AI客服情感胜任力模型和相应的测量指标,认为未来的服务型人工智能应该具备情绪感知、情绪理解和情绪调节能力。该模型和测量指标为今后在技术和管理上发展更具"情商"的服务型人工智能、提高用户对AI客服的满意度提供参考。Artifi cial intelligence(AI)is rapidly being applied in the fi eld of customer service,but many studies have pointed out that users are not usually satisfied with them.It is generally believed that an AI customer service representative(CSR)is not as“understandable”as human CSRs,but there is a lack of research on what kind of“EQ”ability an AI CSR should have.Through the idea of“evaluation from others”,based on the survey data from 757 users,this study constructed a well-fitted competency model and measurement index for the emotional competency which an AI CSR should have(including emotional perception,emotional understanding and emotional regulation).The results provide a reference for developing service-oriented artifi cial intelligence with more“EQ”and improving users'satisfaction with AI customer service in the future.

关 键 词：人工智能 客户服务 客户满意度 情绪智力 情感胜任力 

分 类 号：F273.1[经济管理—企业管理]