Outlier Detection in Discrete Sequences

Abstract

In the previous chapter, the discussion of anomaly detection was focussed on continuous time series. A related problem is the case where the sequences are discrete. In other words, the values at the time stamps are categorical. Such scenarios arise quite commonly in a variety of system diagnosis, intrusion detection and biological data applications. It is to be noted that in some domains such as intrusion detection and system diagnosis, the discrete sequences are caused by temporal ordering, whereas in other domains such as biological data, the discrete sequences are caused by physical ordering.