Authors:
Li Dong
;
Zhang Huan
and
Yu Zitong
Affiliation:
China Electronic Product Reliability and Environmental Testing Research Institute, Dongguanzhuang Road No.110, Guangdong province, China
Keyword(s):
Deep Web Mining, Domain Ontology, Schema Extraction, Query Transformation.
Abstract:
The resources of many Web-accessible databases, which are a very large portion of the structured data on the Web, are only available through query interfaces but are invisible to the traditional search engines. Many methods, which discovery these resources automatically, rely on the different structures of Web pages and various designing modes of databases. However, some semantic meanings and relations are ignored. Here we introduce a Web information retrieval system that obtains the knowledge from multiple databases automatically by using common ontology WordNet. Also, deep Web query results are post-processed based on domain ontology. That is, given an integrated interface, after inputting a query, our system offers an ordered list of data records to users. We have conducted an extensive experimental evaluation of the Web information retrieval system over real documents. Also, we test our system with hundreds of databases on different topics. Experiments show that our system has lo
w cost and achieves high discovering accuracy across multiple databases.
(More)