论文标题
多元数据流的实时异常检测
Real-time Anomaly Detection for Multivariate Data Streams
论文作者
论文摘要
我们提出了基于概率指数加权运动平均值（PEWMA）的数据流的实时多元异常检测算法。我们的公式对数据中的（突然的瞬态，突然的分布和逐渐分布）的弹性具有弹性。新颖的异常检测例程利用增量在线算法来处理流。此外，我们提出的异常检测算法以一种无监督的方式起作用，消除了对示例的需求。我们的算法表现良好，并且在概念漂移面前具有弹性。
We present a real-time multivariate anomaly detection algorithm for data streams based on the Probabilistic Exponentially Weighted Moving Average (PEWMA). Our formulation is resilient to (abrupt transient, abrupt distributional, and gradual distributional) shifts in the data. The novel anomaly detection routines utilize an incremental online algorithm to handle streams. Furthermore, our proposed anomaly detection algorithm works in an unsupervised manner eliminating the need for labeled examples. Our algorithm performs well and is resilient in the face of concept drifts.
