网络跨库检索中基于Ontology的数据抽取与合并  

Data Extraction and Integration in Searching over Multiple Networked Databases

作　　者：薛亮[1] 冯博琴[1] 管涛[1] 

机构地区：[1]西安交通大学电子与信息工程学院,陕西西安710049

出　　处：《小型微型计算机系统》2005年第10期1807-1809,共3页Journal of Chinese Computer Systems

摘　　要：对网络多个信息源跨库检索的结果进行Ontology建模,实现异构分布式数据源的数据抽取与合并．数据抽取首先将 信息源的检索结果页面映射成有限标号树,其次应用抽取规则得到所需数据;给出按库合并算法,使得网络多数据源返回的结 果得以高效合并．实验数据表明将Ontology建模应用于跨库检索结果处理有效而且正确,抽取准确率可以达到100％．An ontology model for the result processing of searching over multiple networked databases （SND） is proposed in the paper, with the goal of data extraction from the result pages returned by the distributed information source and data consolidation. In the model the process of data extraction contains structure of a limited label tree and finding out the useful is given, which consolidates the data from heterogeneous the process of result processing gains good effect and the two parts, mapping the result page of the information source into the data based on the predefined extraction rules. The merging algorithm sources efficiently. Experiment results show that based on ontology precision of extraction reaches 100%.

关 键 词：跨库检索 ONTOLOGY 数据抽取 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]