主要研究了基于HTML文档的信息抽取,提出了一种基于样本实例的Web信息抽取的方法.用户首先选定样本页面和预先定义模式(基于O-R模型),然后对样本页面和其中的样本记录进行标记、学习,形成信息抽取规则,并存入知识库;利用知识库对其他同类页面自动抽取所需的信息,存入数据库中.本方法可用于Web查询,也可用于信息集成的包装器.
[1] 黄豫清, 戚广志, 张福炎, HUANG Yuq-ing, QI Guang-zhi, ZHANG Fu-yan. 从WEB文档中构造半结构化信息的抽取器 [J]. 软件学报 2000.
[2] Wiederhold G.. Mediators in the architecture of future information systems [J]. Computer 1992, 3(3).
[3] MECCA G, ATZENI P. Cut and paste [C]. In Proceedings of 16th ACM SIGMOD Symposium on Principles of Database Systems Tucson Arizona 1997.
[4] Ashish A, KNOBLOCK C. Wrapper generation for semi-structured internet sources [J]. Sigmod Record 1997, 26(04).
[5] HAMMAR J, GARCIA-MOLINA H, CHO J, et al.. Extracting semi-structured information from the web [J]. Sigmod Record 1997, 26(02).
[6] Arnaud Sahuguet, Fabien Azavant. Building intelligent Web applications using lightweight wrappers [J]. Data & knowledge engineering 2001, 3(3).
[7] FLORESCU D, LEVY A, MENDELZON A. Databse techniques for the World-wide web: A Suvey [J]. Sigmod Record 1998, 27(03).
