时间序列不确定数据流中异常数据检测方法

An anomaly detection method for time-series uncertain data streams

结合小波分析和不确定聚类方法的优点,提出一种基时间序列不确定数据流的异常数据检测方法,该方法主要考虑数据流中元组的不确定性,同时平衡检测的计算代价与检测精度。仿真实验证明,该检测方法能够良好地适应数据流的不确定性,在一定条件下可获得相当好的检测效果。

An anomaly detection method for uncertain data streams that combines wavelet analysis and uncertainty clustering is proposed. Taking the uncertainty anomaly detection into account, the minimum computation for reducing the computational cost of detection are estimated at the same time. Experimental results on actual data source show that this method can adapt to the uncertain of data streams well and provide precise instantaneous detection under certain conditions.