基于模糊谱聚类的电力客服智能问答系统  

The power customer service intelligent question answering system based on fuzzy spectral clustering

作　　者：苏立伟 陈海燕 谭火超 方霆 SU Li-wei;CHEN Hai-yan;TAN Huo-chao;FANG Ting(Guangdong Power Grid Co.,Ltd.,Customer Service Center,Guangzhou 510000,China)

机构地区：[1]广东电网有限责任公司客户服务中心,广州510000

出　　处：《信息技术》2023年第1期120-125,共6页Information Technology

摘　　要：为解决电力公司用户服务需求较大与客服数量较少的问题,设计基于模糊谱聚类的电力客服智能问答系统。通过信息辅助模块将问题工单上传至知识库,将完成预处理的问题储存至知识库,通过深度学习分类模型完成问题属性分类后,模糊谱聚类匹配问题答案,计算差异关键词的相似度,查找系列相关问题,通过自助客服模块即时分析后,在系统界面向用户呈现问题答案。实例测试结果表明,该系统具有较低的最坏时间复杂度,可有效加快电力公司的信息传输效率。In order to solve the problem of large user service demand and small number of customer service in power companies, an intelligent question answering system for power customer service based on fuzzy spectral clustering is designed. By uploading the problem work order to the knowledge base through the information assistance module, the preprocessed problems are stored to the knowledge base. The problem attribute classification is completed through deep learning classification model, the problem answers are matched by fuzzy spectrum clustering. The similarity of different keywords is calculated to find a series of related problems, and the answers are presented in the system interface after the timely analysis through the Self-service customer service module. The example test results show that the system has low worst-case time complexity, which can effectively improve the information transmission efficiency of the power company.

关 键 词：模糊谱聚类 电力客服 智能问答系统 相似度计算 关键词匹配 

分 类 号：TM73[电气工程—电力系统及自动化]