首页 | 本学科首页   官方微博 | 高级检索  
     

基于网页结构特征的中文命名实体识别和关联算法
引用本文:任颖,李华伟,吕红. 基于网页结构特征的中文命名实体识别和关联算法[J]. 自动化技术与应用, 2012, 31(1): 28-31
作者姓名:任颖  李华伟  吕红
作者单位:1. 海军航空工程学院,山东烟台,264001
2. 海军航空工程学院,山东烟台264001;山东商务职业学院,山东烟台264001
摘    要:本文针对已有命名实体识别算法在网页结构特征利用方面的问题,提出了基于网页结构特征的中文命名实体识别算法和实体关联算法。该算法结合了网页结构特征,提出了候选实体生成方法,将实体类型识别问题转化为候选实体分类问题。同时提出了基于DOM-Ttee的实体关联算法,实验显示本文的系统是非常有效的。

关 键 词:网顶结构  实体识别  关联算法

Chinese Naming Entity Recognition and Correlation Algorithms Based on Web Strecture Feature
REN Ying , LI Hua-wei , LV Hong. Chinese Naming Entity Recognition and Correlation Algorithms Based on Web Strecture Feature[J]. Techniques of Automation and Applications, 2012, 31(1): 28-31
Authors:REN Ying    LI Hua-wei    LV Hong
Affiliation:1. Naval Aviation Engineering College, Yantai 264001 China; 2. Shandong Businss Institute, Yantai 264001 China )
Abstract:In this paper, has been named entity recognition algorithm is used in the web structure issues, proposes structural features of web-based Chinese named entity recognition algorithms and entities associated with algorithm. The algo- rithm combines the structural features of web pages, generation method proposes candidate entities, the entity type is transformed into the candidate identification entity classification issues. Also proposes based on DOM-Tree of entities associated with algorithm. Experiments show that this system is very effective.
Keywords:web structure  entity recognition  asociation algorithm
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号