[1]张 俊,高志强,徐 惠,等.一种基于Bootstrapping的本体学习方法[J].南京师范大学学报(工程技术版),2008,08(04):056-58.
　Zhang Jun,Gao Zhiqiang,Xu Hui,et al.An Ontology Learning Method Based on Bootstrapping[J].Journal of Nanjing Normal University(Engineering and Technology),2008,08(04):056-58.
点击复制

一种基于Bootstrapping的本体学习方法
分享到：

南京师范大学学报（工程技术版）[ISSN:1006-6977/CN:61-1281/TN]

卷:
08卷
期数:
2008年04期
页码:
056-58
栏目:
出版日期:
2008-12-30

文章信息/Info

Title:
An Ontology Learning Method Based on Bootstrapping
作者:
张 俊1;  高志强1;  徐 惠2;  蔡施彦1;  戴云徽3
1. 东南大学计算机科学与工程学院, 江苏南京210096; 2. 东南大学软件学院, 江苏南京210096;
3. 南京理工大学经济管理学院, 江苏南京210094
Author(s):
Zhang Jun1; Gao Zhiqiang1; Xu Hui2; Cai Shiyan1; Dai Yunhui3
1.College of Computer Science and Engineering,Southeast University,Nanjing 210096,China;2.College of Software Engineering,Southeast University,Nanjing 210096,China;3.College of Economy,Nanjing University of Science and Technology,Nanjing 210094,China
关键词:
信息抽取;  本体学习;  自扩展
Keywords:
in fo rm ation ex traction;  onto logy learn ing;  boo tstrapp ing
分类号:
TP391.1
摘要:
提出了一种基于自扩展的本体学习方法用于获取领域术语.该方法只需提供少量种子术语和一个未标注语料库作为输入,由种子术语开始学习抽取模式,再由学习到的模式发现新的术语,进一步由新发现的术语学习新的抽取模式,如此循环迭代.实验结果表明,该算法能够产生较高质量的领域术语集合和抽取模式集合,这样的集合可用于相关领域的信息抽取.
Abstract:
Th is pape r presents a boo tstrapp ing-based approach fo r on to logy learn ing wh ich can be used in field wo rds acqu isition. This approach only requires a sm a ll se t of seed term s and an unm arked corpus as its input, and it learns extraction patterns using seed te rm s, and detects new fie ld te rm s us ing patte rns lea rned be fo re, and aga in using new ly acqu ired field term s to d iscover nove l ex traction pa tterns, and iterate on. Experim ents show that this approach produces a re lative ly h igh-quality dom a in-specific dictionary and a set of ex traction patternsw hich can be consequently app lied to inform ation extraction in re la ted dom a in

参考文献/References:

[ 1] Du X Y, LiM, W ang S. A survey on onto logy learn ing research[ J]. Journal o f So ftware, 2006, 17( 9): 1 837-1 847.
[ 2] Staab S, H othoA. Sem antic web and m ach ine learn ing[ C] / / Tuto rial at the 22nd Inte rnational Con ference onM achine Learning.Bonn Germ any, Am sterdam: IOS Press, 2005.
[ 3] B lum A, M itche ll T. Comb in ing labe led and unlabe led data w ith co- tra in ing[ C] / / COLT. New Yo rk: ACM Press, 1998: 92-100.
[ 4] R ilo ff E, Jones R. Learn ing d ic tiona ries fo r info rm ation ex traction by m ult-i leve l bootstrapp ing [ C ] / / Proceedings o f the 16 th Na tiona l Conference on Artific ia l Inte lligence. Austin, TX: AAA I Press/TheM IT Press, 1999: 1 044-1 049.
[ 5] R ilo ff E. An emp irical study o f autom ated dictionary construc tion for inform ation ex traction in three dom a ins[ C] / / Artific ial Inte lligence. Karlsruhe: E lsev ier Publishers, 1996( 85) : 101-134.
[ 6] Cunn ingham H, Ga izauskas R, W ilks Y. GATE) ) ) a genera l a rchitec ture for tex t eng ineering[ C] / / Proceed ings o f the 16 th Con ference on Computational L ingu istics. Copenhagen, New Yo rk: ACM Press, 1996.

备注/Memo

备注/Memo:
基金项目: 国家科技计划( 2006BAK10B02)资助项目.
通讯联系人: 高志强, 教授, 博士生导师, 研究方向: 机器学习和本体学习. E-m ail:zqgao@ seu. edu. cn
更新日期/Last Update: 2013-04-24