基于智能的网页信息提取系统的研究与设计  

Research on Web information extraction system based on intelligence and its design

作　　者：刘亚东[1] 彭舰[1] 张达平[1] 

机构地区：[1]四川大学计算机学院,成都610065

出　　处：《四川大学学报（自然科学版）》2009年第4期957-962,共6页Journal of Sichuan University(Natural Science Edition)

基　　金：四川省应用基础基金(2008JY0027);四川省科技支撑计划项目(07GG006-040)

摘　　要：随着Internet的迅速发展,为人们提供了大量的信息,但这些信息都包含在网页中,为了使用这些信息数据,需要将数据从网页中提取出来.本文介绍了一种新的基于智能的网页信息提取系统EIES,通过对RoadRunner的改进和利用,在提取过程中不需人工干预,实现了信息提取的智能化.实验表明,该系统能够更准确、更有效地分类相似网页和提取网页信息.Along with the Internet rapid development, a mass of information is supplied for people, but all these information is in the Web pages. In order to make use of these information data, it＇s needed to extract the data from the pages. This paper presents a new Web information extraction system based on intelligence-EIES. Through improving and using RoadRunner, it realizes the information extraction intellectualization without any manual work in the extraction process. The experiment indicates that this system can be more accurate and more effective to classify similar pages and extract web information.

关 键 词：智能 WEB信息提取 RoadRunner HTML结构树 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]