The Community for Technology Leaders
Green Image
Issue No. 05 - May (2011 vol. 23)
ISSN: 1041-4347
pp: 774-787
Dominik Fisch , University of Passau, Passau
Thiemo Gruber , University of Passau, Passau
Bernhard Sick , University of Passau, Passau
In this article, we provide a new technique for temporal data mining which is based on classification rules that can easily be understood by human domain experts. Basically, time series are decomposed into short segments, and short-term trends of the time series within the segments (e.g., average, slope, and curvature) are described by means of polynomial models. Then, the classifiers assess short sequences of trends in subsequent segments with their rule premises. The conclusions gradually assign an input to a class. As the classifier is a generative model of the processes from which the time series are assumed to originate, anomalies can be detected, too. Segmentation and piecewise polynomial modeling are done extremely fast in only one pass over the time series. Thus, the approach is applicable to problems with harsh timing constraints. We lay the theoretical foundations for this classifier, including a new distance measure for time series and a new technique to construct a dynamic classifier from a static one, and demonstrate its properties by means of various benchmark time series, for example, Lorenz attractor time series, energy consumption in a building, or ECG data.
Temporal data mining, time series classification, anomaly detection, piecewise polynomial representation, piecewise probabilistic representation, generative classifier, SwiftRule.

B. Sick, T. Gruber and D. Fisch, "SwiftRule: Mining Comprehensible Classification Rules for Time Series Analysis," in IEEE Transactions on Knowledge & Data Engineering, vol. 23, no. , pp. 774-787, 2010.
86 ms
(Ver 3.3 (11022016))