关键词: 数据流 数据挖掘 事件检测 网格聚类

Abstract:

Event detection is one of the most important issues in event processing systems. Outlier, change and burst are three typical types of events. We address how to detect multiple types of events simultaneously. In this paper,we first explore the relationship of these three types of events, and then present a unified method for dealing with all of them by using grid-based clustering. In order to evaluate the events, several score functions are defined for each type of events as well. Simulation results testify the efficiency of the proposed framework.

Key words: data stream, data mining;event detection, grid-based clustering