Mining Sequence Pattern from Time Series Based on Inter-relevant Successive Trees Model
暂无分享,去创建一个
In this paper, a novel method is proposed to discover frequent pattern from time series. It first segments time series based on perceptually important points, then converted it into meaningful symbol sequences by the relative scope, finally used a new mining model to find frequent patterns. Compared with the previous methods, the method is simpler and more efficient.
[1] Heikki Mannila,et al. Rule Discovery from Time Series , 1998, KDD.
[2] Heikki Mannila,et al. Discovering Generalized Episodes Using Minimal Occurrences , 1996, KDD.
[3] Eamonn J. Keogh,et al. A Probabilistic Approach to Fast Pattern Matching in Time Series Databases , 1997, KDD.