An Algorithm for Outlier Detection on Uncertain Data Stream

Abstract

Outlier detection plays an important role in fraud detection, sensor net, computer network management and many other areas. Now the flow property and uncertainty of data are more and more apparent, outlier detection on uncertain data stream has become a new research topic. Firstly, we propose a new outlier concept on uncertain data stream based on possible worlds. Then an outlier detection method on uncertain data stream is proposed to meet the demand of limited storage and real-time processing. Next, a dynamic storage structure is designed for outlier detection on uncertain data stream over sliding window, to meet the demands of limited storage and real-time response. Furthermore, an efficient range query method based on SM-tree(Statistics M-tree) is proposed to reduce some redundant calculation. Finally, the performance of our method is verified through a large number of simulation experiments. The experimental results show that our method is an effective way to solve the problem of outlier detection on uncertain data stream, and it could significantly reduce the execution time and storage space.

This research are supported by the NSFC (Grant No. 61173029, 61025007, 60933001, 75105487 and 61100024), National Basic Research Program of China (973, Grant No. 2011CB302200-G), National High Technology Research and Development 863 Program of China (Grant No. 2012AA011004) and the Fundamental Research Funds for the Central Universities (Grant No. N110404011).