基于自然语言处理的意图驱动网络表征
Intent-driven network representation based on natural language processing
姬泽阳 1杨春刚 1李富强 2欧阳颖 3刘祥林3
作者信息
- 1. 西安电子科技大学通信工程学院,陕西西安 710071;西安电子科技大学杭州研究院,浙江杭州 311200
- 2. 中国电子科技集团公司数据链重点实验室,陕西西安 710068
- 3. 西安电子科技大学通信工程学院,陕西西安 710071
- 折叠
摘要
巨量网络规模、复杂网络结构和低效人工配置等问题需要网络配置自动化和无人化.意图驱动网络可实现去人工的网络自动配置,其中意图表征是关键;然而,现有意图表征范式未能形成统一的标准语法规则.提出了自然语言处理与知识图谱相结合的意图驱动网络表征系统,该系统支持语音、文本等形式的意图输入.所提意图表征方法利用文本检错、纠错和相似度检测技术提升了意图表征的效果,将意图表征结果保存为知识图谱,实现了标准、统一的语法规则,并通过实验验证了该系统的有效性.
Abstract
Problems such as massive network scale,complex network structure,and inefficient manual configuration require automated and unmanned network configuration.Intent-driven networks can realize automatic configuration of the network without human labor,where intent representation is the key.However,the existing intent representation paradigm fails to form a uniform standard syntax rule.An intent-driven network representation system based on the combination of natural language processing and knowledge graph is proposed,which supports intentional input in the form of speech and text.The proposed intent representation method uses text error detection,error correction and similarity detection technology to achieve the effect of improving intent representation,saves the intent representation results as a knowledge graph,and realizes standard and unified syntax rules.Finally,the effectiveness of the system is verified by experiments.
关键词
意图驱动网络/意图表征/自然语言处理/知识图谱Key words
intent-driven network/intent representation/natural language processing/knowledge graph引用本文复制引用
基金项目
数据链技术重点实验室开放基金(CLDL-20202314)
出版年
2024
NETL
NSTL
万方数据