A easy user interface of IR system over large scale deep web

Abstract

The Deep Web is characterized by its large scale, domain diversity, very high-quality content, and vast heterogeneous collections of structured data. Traditional data integration techniques are no longer valid in the face of such heterogeneity and scale, and the integrated interfaces can become too complicated for ordinary users to use. In this paper, we propose a novel keyword-based user interface system EasyUI for achieving web-scale data integration and easy to use for ordinary users. Dealing with heterogeneity on the web-scale presents many new challenges. We proposed new methods to address these challenges, i.e., indexing schemata terms, data values, and domains features of the Deep Web, processing user input, mapping the user query to domain that the user query most likely corresponds to, and translating the user query into candidate structured queries for the domain. Our experiments show that EasyUI performs effectively.