Research progress on outlier mining
Dongqing Yang · Caai Transactions on Intelligent Systems · 2006
An outlier is a data point that is significantly different from the others in a data set. One person’s noise could be another person’s signal, and therefore the problem of outlier mining attracts more and more interests in research of information science when the research fields of data quality, fraud detection, intrusion detection, fault diagnosis, military scout and so on receive wide attentions. In this paper, a survey was presented for the problem of outlier mining from the basic concepts to the principal research problems and the underlying techniques, including origination of outlier, definition of outlier and the comparison of popular outlier mining methods. A summary of the current state of the art of these techniques, a discussion on future research topics, and the challenges of the outlier mining were also presented.