华南理工大学学报(自然科学版) ›› 2025, Vol. 53 ›› Issue (3): 40-49.doi: 10.12141/j.issn.1000-565X.240191

• 计算机科学与技术 • 上一篇    下一篇

基于意图理解驱动的客服知识推荐大模型构建

马晓亮1,2,3(), 高洁2,3, 刘英2,3, 裴庆祺1, 赵汝强2,4, 杨邦兴5, 邓从健3,4   

  1. 1.西安电子科技大学 广州研究院,广东 广州 510555
    2.中国电信股份有限公司 广州分公司,广东 广州 510620
    3.马晓亮劳模和创新工匠工作室,广东 广州 510620
    4.广州云趣信息科技有限公司,广东 广州 510665
    5.中数通信息有限公司,广东 广州 510650
  • 收稿日期:2024-04-17 出版日期:2025-03-10 发布日期:2024-08-23
  • 作者简介:马晓亮(1973—),男,博士生,高级工程师,华南理工大学工商管理学院讲席教授,主要从事人工智能、自然语言处理、方言处理、运营商客服运营、数据安全保护研究。E-mail: maxiaol.gd@chinatelecom.cn
  • 基金资助:
    国家重点研发计划项目(2021YFB2700600)

Customer Service Knowledge Recommendation Large Model Construction Driven by Intent Understanding

MA Xiaoliang1,2,3(), GAO Jie2,3, LIU Ying2,3, PEI Qingqi1, ZHAO Ruqiang2,4, YANG Bangxing5, DENG Congjian3,4   

  1. 1.Guangzhou Institue of Technology,Xidian University,Guangzhou 510555,Guangdong,China
    2.Guangzhou Branch of China Telecom Co. ,Ltd. ,Guangzhou 510620,Guangdong,China
    3.MA Xiaoliang’?s Model Worker and Innovative Craftsman Workshop,Guangzhou 510620,Guangdong,China
    4.Guangzhou Yunqu Information Technology Co. ,Ltd. ,Guangzhou 510665,Guangdong,China
    5.China DataCom Co. ,Ltd. ,Guangzhou 510650,Guangdong,China
  • Received:2024-04-17 Online:2025-03-10 Published:2024-08-23
  • Supported by:
    the National Key R & D Program of China(2021YFB2700600)

摘要:

随着人工智能技术在客服领域的深入应用,电信运营商对AI服务知识推荐的准确率提升提出了更高的要求。为提高电信运营商AI客服系统的知识推荐效率和准确度,该文提出了基于意图理解驱动的客服知识推荐大模型。首先,采用同义词及对话序列的关键词提取模型识别用户查询中的关键词,通过语义相似度比较技术匹配标准问库中的问题,生成最相关的标准问,并采用生成式智能体技术框架构建标准问库,使用智能体技术自动生成知识问题;然后将提取的标准问输入ChatGLM2-6B大语言模型中,经过预训练与人类偏好对齐训练,以进一步提高知识推荐的准确率。实验结果显示:引入标准问库后,智能推荐系统在特定行业知识领域的准确率从74.8%显著提升至85.9%,多组对比实验结果进一步验证了建立标准问库的策略在提高准确率方面的有效性;该文大模型优化了运营商AI客服的智能知识推荐,可为电信运营商AI客服系统的知识推荐提供新的思路和技术支持;运营商通过该文大模型能够更有效地理解和响应客户查询,显著提升客户服务体验。

关键词: 意图理解, AI客服, 生成式大语言模型, 关键词提问, 提示示例

Abstract:

With the deepening application of artificial intelligence technology in the field of customer service, telecommunications operators have raised higher standards for the accuracy of AI service knowledge recommendations. To enhance the efficiency and accuracy of knowledge recommendation in telecommunications operators’AI customer service systems, this paper proposed a large-scale customer service knowledge recommendation model driven by intent understanding. Firstly, the synonym and dialogue sequence keyword extraction model was employed to identify key terms in user queries. These keywords were then matched with questions in a standard question bank using semantic similarity comparison techniques to generate the most relevant standard questions. Additionally, a generative agent technology framework was utilized to construct and enrich the standard question bank, enabling the automatic generation of knowledge questions. The extracted standard questions were input into the ChatGLM2-6B large language model, which has been pre-trained and aligned with human preferences, further improving the accuracy of knowledge recommendations. The experimental results show that after the introduction of the standard question bank, the accuracy of the intelligent recommendation system in specific industry knowledge domains significantly increased from 74.8% to 85.9%. Multiple sets of comparative experimental results further validate the effectiveness of the strategy of establishing a standard question bank in improving accuracy. The large model discussed in this paper optimized the intelligent knowledge recommendation for operator AI customer service, providing new ideas and technical support for the knowledge recommendation in telecommunications operators’AI customer service systems. With this model, operators can more effectively understand and respond to customer inquiries, significantly enhancing the customer service experience.

Key words: intent understanding, AI customer service, large language model, keyword-based questioning, prompt example

中图分类号: