改进的基于本体的Web信息抽取  

IMPROVED ONTOLOGY-BASED WEB INFORMATION EXTRACTION

作　　者：陈雪[1] 梁永全[1] 赵相彬[1] 

机构地区：[1]山东科技大学信息科学与工程学院,山东青岛266590

出　　处：《计算机应用与软件》2013年第7期14-16,42,共4页Computer Applications and Software

基　　金：国家自然科学基金专项基金项目(71240003);山东省自然科学基金项目(ZR2012FM003);山东科技大学专项基金项目(YCA120341;YCA120345)

摘　　要：针对Web信息抽取主要依据内容进行抽取的思想,通过对基于本体的Web信息抽取技术的分析,在对网页进行预处理的过程中引入网页分块思想。通过网页分块技术准确提取出所需信息的抽取区域,再根据本体抽取规则对得到的抽取区域进行处理,最终得到需要的信息。实验验证了改进的基于本体的Web信息抽取算法具有更高的准确率和召回率,具有良好的应用前景。In light of the idea that the Web information extraction mainly to be done according to the content, by analysing the ontology- based Web information extraction technology, we introduce webpage segmentation concept in the process of webpage pretreatment. Through the webpage segmentation technology we accurately extract the extraction region of the required information, then process it according to ontology extraction rules, and ultimately get the information required. Through the experiments we verify that the improved ontology-based Web information extraction algorithm has higher precision and recall rate, and has good prospects in practical applications.

关 键 词：信息抽取 网页分块 本体 抽取规则 准确率 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]