吕诗宁,张毅,胡若云,沈然,江俊军,欧智坚
LYU Shining,ZHANG Yi,HU Ruoyun,SHEN Ran,JIANG Junjun,OU Zhijian

• 1:国网浙江省电力有限公司电力科学研究院
• 2:清华大学电子工程系
• 3:国网浙江玉环县供电有限公司

摘要(Abstract)：

为了开发面向电力领域的智能客服对话系统,提出了融合神经网络和电力领域知识的方法。利用基于神经网络的自然语言理解技术从客户的话中提取有用信息;利用对话流配置框架和知识图谱将电力客服业务流程、客服数据库和标准问答对等电力领域知识融入对话系统,指导对话系统给出正确的回复。最后通过实现两个电力领域客服对话系统验证了方法的可行性。
The paper presents a method integrating neural networks and electric power field knowledge to develop an intelligent customer service dialogue system oriented to the electric power field. Natural language understanding technology based on neural networks is used to extract useful information from customers′ utterances; the dialogue flow configuration framework and knowledge graph are used to integrate field knowledge of electric power such as the customer service business process, customer service database and standard Q&A into the dialogue system to guide the dialogue system to give correct replies. Finally, the feasibility of the method is verified by implementing two customer service dialogue systems in the electric power field.

关键词(KeyWords)： 对话系统;神经网络;电力领域知识;智能客服
dialogue system;neural network;electric power field knowledge;intelligent customer service

Abstract:

Keywords:

基金项目(Foundation): 国家电网有限公司科技项目（5211DS180030）

作者(Author): 吕诗宁,张毅,胡若云,沈然,江俊军,欧智坚
LYU Shining,ZHANG Yi,HU Ruoyun,SHEN Ran,JIANG Junjun,OU Zhijian

DOI: 10.19585/j.zjdl.202008012

参考文献(References)：

• [1]丁麒，张爽.杭州电力客服95598热线优化排班分析[J].浙江电力，2008,27(6):52-55.
• [2]吴侃侃.OA系统与客服中心集成的实现[J].浙江电力，2004(2):65-66.
• [3]张千福，黄晓光，曹璐，等.电力智能客服实现的技术研究[J].数字通信世界，2019(5):121-122.
• [4]吴瀚文.新时代电力企业智能客服应用研究与展望[J].科学技术创新，2018(18):67-68.
• [5]饶竹一，张云翔.基于知识图谱的智能客服系统研究[J].电力信息与通信技术，2017,15(7):41-45.
• [6]GAO J F,GALLEY M,LI L H.Neural approaches to conversational AI[C]//proceedings of ACL 2018,tutorial abstracts,Melbourne,Australia.Stroudsburg,PA,USA:Association for computational Linguistics,2018:2-7.
• [7]MRKSIC'N，O SEAGHDHA D,WEN T-H,et al.Neural belief tracker:data-driven dialogue state tracking[C]//Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers),Vancouver,Canada.Stroudsburg,PA,USA:Association for Computational Linguistics,2017:1777-1788.
• [8]MRKSIC'N,VULIC'I.Fully Statistical Neural Belief Tracking[C]//Proceedings of the 56th Annual Meeting of the As sociation for Computational Linguistics(Volume 2:Short Papers),Melbourne,Australia.Stroudsburg,PA,USA:Association for Computational Linguistics,2018:108-113.
• [9]DAO Y P,OU Z J,REN D W,et al.Tracking of enriched dialog states for flexible conversational information access[EB/OL].(2017-11-09)[2020-04-18].https://arxiv.org/abs/1711.03381.
• [10]NGUYEN D Q.An overview of embedding models of entities and relationships for knowledge base completion[EB/OL].(2017-05-23)[2020-04-18].https://arxiv.org/abs/1703.08098.
• [11]SEO M,KEMBHAVI A,FARHADI A,et al.Bidirectional Attention Flow for Machine Comprehension[EB/OL].(2016-11-05)[2020-04-18].https://arxiv.org/abs/1611.01603.
• [12]LIU X,SHEN Y,DUH K,et al.Stochastic answer net works for machine reading comprehension[C]//Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics(Volume 1:Long Papers),Melbourne,Australia.Stroudsburg,PA,USA:Association for Computational Linguistics,2018:1694-1704.
• [13]RADFORD A,NARASIMHAN K,SALIMANS T,et al.Improving language understanding with unsuper-vised learning[EB/OL].(2018-06-11)[2020-04-18].https://opencui.com/blog/langnage-unsupervised/.
• [14]DEVLIN J,CHANG M W,LEE K,et al.Bert:Pre-training of deep bidirectional transformers for language understanding[EB/OL](2018-10-11)[2020-04-18].https://arxiv.org/abs/1810.04805.
• [15]HENDRYCKS D,GIMPE K.Bridging nonlinearities and stochastic regularizers with gaussian error linear units[EB/OL].(2016-06-02)[2020-04-18].https://arxiv.org/abs/1606.08415.
• [16]BA J L,KIROS J R,HINTON G E.Layer normalization[EB/OL].(2016-07-21)[2020-04-18].https://arxiv.org/abs/1607.06450.
• [17]LAN Z,CHEN M,GOODMAN S,et al.Albert:A lite bert for self-supervised learning of language representations[EB/OL].(2019-09-26)[2020-04-18].http://arxiv.orglabs/1909.11942.