Abstract
In this paper, we present our initial research in building a document retrieval system with following main features: processing English queries of users, interacting with users to correct incorrect syntax queries, giving results of the queries. One of the striking points of this system model is introducing a semantic rendition for natural language queries. Our research is limited in some specific applications searching ebooks in e-libraries with some information about ebooks. In these applications, information about application fields, data structures... must be clearly understood.