点击流数据仓库中基于事件驱动的星型ER模型

Web应用产生了大量的点击流数据,为有效分析这些数据,使之发挥更好的决策支持作用,产生了点击流数据仓库。较之传统数据仓库,点击流数据仓库有更复杂的体系结构及更丰富的数据源,现有的数据模型已不再适合点击流数据仓库的发展需要。提出了一种基于事件驱动的星型ER模型作为点击流数据仓库的概念模型,给出了关键事件/非关键事件及主动关注期望/被动关注期望概念。分析结果表明,该模型克服了现有模型的一些不足,能够很好地适应点击流数据仓库的体系结构。

StarER model based on event drive for clickstream data warehouse

A mount of clickstream data are created by application of Web,in order to analyze data efficiently to make full use of decision support,clickstream data warehouse emerges.Compare to traditional data warehouse,clickstream data warehouse has a more complicated architecture and more abundant data sources,the current data model doesn’t satisfy the needs of clickstream data warehouse’s development any more.This paper presents an event drive starER model as a clickstream data warehouse conceptual model base on above,proposes definitions: critic event/non critic event,active/passive attention anticipation.The analysis results demonstrate that the model overcomes some drawbacks from other models,which is adaptive for the architecture of clickstream data warehouse.