Time Series Data Mining

Data Mining or Knowledge Discovery in Databases (KDD) is an important area of computer sciences. The relevance of this area is due to the enormous quantity of information daily produced by different sources, for instance the web, biological processes, finance, the aeronautic industry, retail, and telecommunications data. A considerable amount of this information represents temporal events which are typically stored in the form of time series. There are several phenomena expected to be identified among databases of this type, namely through motif (pattern) discovery, classification, clustering, query by content, abnormality detection, and forecast of property values. We focus particularly on the area of time series motif discovery (Lin and Keogh 2002) , also known as the extraction of recurrent patterns. These patterns are relevant because they summarise the time series of a domain and help the domain expert understand the database at hand (Ferreira et al. 2006). Figure 1 shows one example of such type of pattern in the context of electroencephalogram (EEG) time series. This specific motif is detected in three different time series in the database.