基于知识图谱的智能语音客服系统设计

Design of Intelligent Voice Customer Service System Based on Knowledge Graph

摘要: 本文介绍了主流的智能客服实现方案，并提出了一种基于知识图谱的语音识别智能客服系统的设计方案。通过语音识别的方式获取用户提出的问题内容，利用各项自然语言处理技术对问题进行建模并提取关键信息，结合该领域的知识图谱，检索出符合要求的问题答案，最后把消除歧义的最终结果反馈给用户，完成问答交互。能够有效削减人工传统客服的各类成本开销，并利用智能客服系统提高工作效率。

Abstract: This article introduces mainstream smart customer service implementation schemes, and proposes a design scheme of a voice recognition intelligent customer service system based on knowledge graph. Voice recognition is used to obtain the content of the questions raised by users, various natural language processing technologies are used to model the questions and extract key information, the knowledge map is combined in the field, and the answer is retrieved to the questions that meet the requirements. The results are fed back to the user to complete the Q & A interaction. The cost of manual customer service is reduced and the efficiency of the customer service system is improved.