Title :
Computer anomaly detection based on the moving averages of the power series distributed random sequence
Author_Institution :
Dept. of Inf. Sci. & Technol., East China Univ. of Political Sci. & Law, Shanghai, China
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.
Keywords :
Poisson distribution; data analysis; entropy; moving average processes; random sequences; security of data; series (mathematics); MREDD; Poisson distribution model; anomaly detection model distribution; binomial distribution model; computer anomaly detection; data analysis; data comparison; moving relative entropy density deviation method; negative binomial model; power series distributed random sequence; Computational modeling; Computers; Data models; Educational institutions; Entropy; Intrusion detection; Random variables; anomaly detection; distribution of power series; moving average; moving relative entropy density deviation;
Conference_Titel :
Fuzzy Systems and Knowledge Discovery (FSKD), 2014 11th International Conference on
Conference_Location :
Xiamen
Print_ISBN :
978-1-4799-5147-5
DOI :
10.1109/FSKD.2014.6980852