knowledge-based HTML document generation for utomating web publishing

摘要

Among the dozen most popular applications on the Internet, the World Wide Web (WWW) is showing the fastest growth. Its power comes from a simple but powerful documentation language, HyperText Markup Language. In this paper we propose a knowledge-based approach and schema to facilitate the creation and maintenance of HTML-based documentation for a set of documents whose content is partially derived from other information souses, such as databases. We denote the system as the Knowledge-Based HTML Document Generator (KHDG). The basic idea is to use regularly updated information on information servers, such as databases, as the major source of knowledge for KHDG. Analyzing and authoring expertise are employed as rules for reasoning about the construction of HTML documents and their interrelationships. The index information about reusable document components is maintained in the Document Catalog Base (DCB). This scheme makes the components stored in distributed repository consistent and easy to integrate. KHDG produces the HTML document in a threestep process: source information analysis and extraction, HTML document planning and HTML page generation. The implementation of the system is based on CGI scripting in a Windows web server. We suggest that the approach utilized in KHDG is capable of automating routine HTML document generation and maintenance based on regularly updated databases.