ARML

Abstract

As database systems are applied to a wide range of domains associated with highly complicated information processing and the information management paradigm shifts from data management to knowledge management, an active support for various information services has become a practical approach to add intelligence to the existing database support. The active support complements auto-everything technology that minimizes users’ intervention in the course of decision-making. Typically, active information management has a rule-processing feature based on an Event-Condition-Action (ECA) construct by which complicated business strategies can be modeled and processed without looking into application programs and the underlying database. ECA-based active rules specify that certain actions should be executed automatically whenever certain events occur and specified conditions are met. By encapsulating the business logic of an enterprise into active rules, the changes of business strategy can be made by modifying just the relevant rules, not the whole application. Different active systems use different rule definition languages for rule description, and they have their own syntactic and semantic features thus making rule reusability and knowledge sharing difficult. In this thesis, we propose an XML-based ECA rule definition language, called Active Rule Markup Language (ARML) to make active rules interoperable among different applications and systems. With its easy-to-define rule description facility using XML, ARML provides a uniform ECA rule definition construct for heterogeneous active information systems allowing sharing of rules among different applications and exchanging knowledge between business partners. Taking advantages of XML, such as standard markup language with its ease of understanding, excellent expressiveness, and web representation power, ARML is simple to use and easy to implement by using well developed XML APIs and XML utilities. ARML defines a new ta...