基于感知机器共情的人工智能客服交互过程设计与用户满意度提升

AI聊天机器人降低了客服难度,但是却存在服务满意度始终低于人工服务的难题。本课题融合了信息系统研究中的行为科学和设计科学方法，提出了一种新的AI交互过程设计思路：基于“机器共情”的顺序问答交互过程（eFAQ）：通过机器学习发现全体用户AI对话模式，然后结合当前用户问题预测其意图，并根据匹配程度给出不同的确认交互；结合交互界面的拟人化设计，在认知和情感上让用户对AI产生机器共情的感知，提高用户满意度。在给出具体的AI交互过程设计之前，本课题借鉴共情心理机制，通过神经营销学/神经信息系统的方法，验证AI界面和交互方式对感知机器共情的影响。在给出交互设计之后，又进一步运用大数据自然实验的方法，验证因果效应。.机器学习算法、信息系统设计、用户行为，过去是三个相互独立的研究领域。本研究基于设计科学的思路，以“感知机器共情”为机制核心，并通过机器学习的方法加以实现，是交叉方法融合的探索。

AI chatbots reduce the difficulty of customer service, but there is a problem that service satisfaction is always lower than manual services. This topic integrates the behavioral science and design science methods in the research of information systems, and proposes a new AI interactive process design idea: a sequential question answering interactive process based on "machine empathy" (eFAQ): discovering AI for all users through machine learning Dialogue mode, then predict the current user's subsequent possible questions, and give confirmation prompts; combined with the anthropomorphic design of the interactive interface, cognitively and emotionally allow users to generate AI empathy perception of AI, thereby improving user satisfaction. Before giving a specific design of the AI ​​interaction process, this subject draws on the psychological mechanism of empathy and uses Neuromarketing / Neural IS methods to verify the influence of the AI ​​interface and interaction on the perception of empathy in the machine. After giving the interactive design, the method of big data natural experiment is further used to verify the causal effect..In the past, machine learning algorithms, information system design, and user behavior were three independent research areas. This research is based on the idea of ​​design science, with "perceived machine empathy" as the core of the mechanism, and the interactive process in cognitive empathy is implemented through machine learning methods, which is an exploration of cross-method fusion.