Application-oriented document processing supported by knowledge database

Computer-automatic or computer-aided document generation is highly desired in applications. Because it can implement standardized document management, establish new workflows and working modes, improve work efficiency and accelerate the process of enterprise information. In this paper a framework, with a new document structure description language - DSDL language, a syntax model and a semantic model, for application-oriented document processing supported by knowledge database is proposed. Using this framework, we have implemented a prototype system and successfully run many examples. The work in this paper also uses machine learning method to provide most of the information for document generation.