Managing Very Large Document Collections Using Semantics

In this paper, a system is presented where documents are no longeridentified by their file names. Instead, a document is represented byits semantics in terms of descriptor and content vector.The descriptor of a document consists of a set of attributes,such as date of creation, its type, its size, annotations, etc. Thecontent vector of a document consists of a set of terms extractedfrom the document. In this paper, a semantic document management systemXBASE is designed and implemented based on the semantics and thefunctions of three main modules, X-Loader, X-Explorer and X-Query.