Computer anomaly detection based on the moving averages of the power series distributed random sequence

Abstract

In order to quickly determine the distribution of anomaly detection model based on small amounts of collected data, the moving relative entropy density deviation method (MREDD) is introduced to test the power series distributed random sequence. Through the moving averages of data analysis and comparison, the anomaly detection models can quickly be established. Experimental results show that this method can be used not only to adaptively choose from the negative binomial model, the binomial distribution model and the Poisson distribution model, but also to reduce the false alarm rate.