Active Information Service Based On ECA Rule In Telecom Data Warehouse

Data Warehouse technology is widely applied in the telecom industry, and it has been used for middle and upper classes of the enterprise to better support business analysis and decisions. However, it supports litter to day-to-day tactical decision, refer to critical event, the fresh data can not be loaded to data warehouse in time, but it can not initiatively push the information to the appropriate users. To solve these problems, active real-time data warehouse is interested by both industry and the research communities. It reduces the time span between critical business events and the actions taken as a result of such events.In this paper, we introduce the Event-Condition-Action (ECA) Rules in Active Database to the conventional data warehouse and propose an ECA based framework for active information service system. It appends two loose-coupling modules: real-time data updating module and active information service module. We propose an object-oriented method to describe ECA rule, and supply APIs like nature language for business users to easily define and understand their business rules based on the business objects, and combine java rule engine with JMS to implement to the active information service module. Finally we validate the effectiveness of the prototype system by a customer churn analysis application in the field of telecommunication.