Extensible rule markup language (XRML) toward Intelligent web platform : design principles and applications

Abstract

To propose the framework of integration architecture between the Knowledge Based System (KBS) and web-based Knowledge Management System (KMS), we design an Extensible Rule Markup Language (XRML). It can support the automatic processing of implicit rules embedded in hypertext as well as browsing them for human comprehension. XRML aims to convert it to XML for browsing, and assist the generation and maintenance of consistent explicit rule structure. To meet this end, XRML requires three components: Rule Identification Markup Language (RIML), Rule Structure Markup Language (RSML), and Rule Triggering Markup Language (RTML). RIML identifies the existence of implicit rules in hypertexts, and specifies its formal association with the explicit structured rules. RSML is an intermediate rule representation which can be automatically transformed to structured rules, and which can be semi-automatically generated and maintained from the hypertext. RTML is a language that defines the conditions when certain rules will be triggered. To design the XRML, we identify six design criteria to be pursued: Expressional Completeness; Relevance Linkability; Polymorphous Consistency; Applicative Universality; Knowledge Integrability; and Interoperability. To implement above ideal goals, XRML is a framework of integrating KBS and KMS. By adopting the framework of XRML, the huge pool of knowledge on the Web sites can be selectively processed for important applications such as automated form processing, preventive auditing, and agent-based electronic commerce. The thesis explained the concept with illustrations, and identified the application and research potentials. This framework can converge the knowledge-based system in AI with Web-based knowledge management system as the next generation of semantic Web platform. Based on this framework, we design and implement a prototype system named Form/XRML in which RTML embedded in form triggers the KBS, which in turn performs inference. R...