摘要
传统自然语言中歧义字段切分系统设计对于歧义字段的分辨能力较差,切分效率差,准确度低。针对上述问题,设计一种基于知识图谱的自然语言中歧义字段切分系统。系统硬件设计了三个模块:采集及分词知识提取模块负责对自然语言中的字段进行收集与信息提取,辨别歧义字段;算法与测试模块处理负责检测所捕捉字段的歧义字段信息,提高系统精准度;分词识别模块负责对歧义字段进行系统切分。软件设计了系统的各项功能,包括系统分词精度提升功能、速度提升功能、完备性增强功能、可维护性以及系统可移植性增强功能,综合整理各结构的性能,进一步提高整体系统切分能力,以实现对歧义字段的切分目的。为检测系统工作效果,与传统系统进行实验对比,结果表明,基于知识图谱的自然语言中歧义字段切分系统设计的切分效果优于传统系统设计。
The segmentation system of ambiguous field in traditional natural language has poor resolution ability,poor segmentation efficiency and low accuracy.To solve these problems,a knowledge map based segmentation system of ambiguity field in natural language is designed.Three modules are designed for the system hardware.The acquisition and segmentation knowledge extraction module is responsible for collecting and extracting the information from fields in natural language,and distinguishing the ambiguous fields.The algorithm and testing module is responsible for detecting the ambiguous field information of captured fields,and improving the accuracy of the system.The segmentation recognition module is responsible for segmenting the ambiguous fields.The various functions are designed for the system software,including the functions of accuracy improvement,speed⁃up,and completeness,maintainability and portability enhancement.The performance of each structure is also integrated and the ability of the whole system to segment ambiguous fields is further improved.In order to detect the working effect of the system,some comparative experiments for the system are carried out in combination with traditional systems.The results show that the design of ambiguity field segmentation system based on knowledge map is better than that of the traditional system.
作者
杨凡
任丹
YANG Fan;REN Dan(School of Computer Engineering,Hubei University of Arts and Science,Xiangyang 441053,China)
出处
《现代电子技术》
北大核心
2020年第1期44-47,52,共5页
Modern Electronics Technique
基金
国家语委科研项目（YB135-109）
关键词
知识图谱
自然语言
歧义字段切分
系统设计
信息提取
效果检测
knowledge map
natural language
ambiguity field segmentation
system design
information extraction
effect detection
