Pattern Mining for Time Series Based on Cloud Theory Pan-concept-tree

  • Yingjun Weng
  • Zhongying Zhu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3066)


One important series mining problems is finding important patterns in larger time series sets. Two limitations of previous works were the poor scalability and the robustness to noise. Here we introduce a algorithm using symbolic mapping based on concept tree. The slope of subsequence is chosen to describe series data. Then, the numerical data is transformed into low dimension symbol by cloud models. Due to characteristic of the cloud models, the loss of data in the course of linear preprocessing is treated. Moreover, it is more flexible for the local noise. Second, cloud Boolean calculation is realized to automatically produce the basic concepts as the leaf nodes in pan-concept-tree which leads to hierarchal discovering of the knowledge .Last, the probabilistic project algorithm was adapted so that comparison among symbols may be carried out with less CPU computing time. Experiments show strong robustness and less time and space complexity.


Pattern Mining Dynamic Time Warping Cloud Model Concept Hierarchy Matching Matrix 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Hegland, M., Clarke, W., Kahn, M.: Mining the MACHO dataset. Computer Physics Communications 142(1-3), 22–28 (2002)CrossRefGoogle Scholar
  2. 2.
    Engelhardt, B., Chien, S., Mutz, D.: Hypothesis generation strategies for adaptive problem solving. In: Proceedings of the IEEE Aerospace Conference, Big Sky, MT (2000)Google Scholar
  3. 3.
    Tompa, M., Buhler, J.: Finding motifs using random projections. In: Proceedings of the 5th Int’l Conference on Computational Molecular Biology, Montreal, Canada, pp. 67–74 (2001)Google Scholar
  4. 4.
    Keogh, E., Chakrabarti, K., Pazzani, M., et al.: Dimensionality reduction for fast similarity search in large time series databases. Journal of Knowledge and Information Systems 3(3), 263–286 (2000)CrossRefGoogle Scholar
  5. 5.
    Li, D.Y., Cheung, D., Shi, X.M., et al.: Uncertainty reasoning based on cloud models in controllers. Computer Math. Applic. 35(3), 99–123 (1998)zbMATHCrossRefGoogle Scholar
  6. 6.
    Weng, Y.J., Zhu, Z.Y.: Research on Time Series Data Mining Based on Linguistic Concept Tree Technique. In: Proceeding of the IEEE Int’l Conference on Systems, Man & Cybernetics, Washington, D.C., pp. 1429–1434 (2003)Google Scholar
  7. 7.
    Jiang, R., Li, D.Y.: Similarity search based on shape representation in time-series data sets. Journal of computer research & development 37(5), 601–608 (2000)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Yingjun Weng
    • 1
  • Zhongying Zhu
    • 1
  1. 1.Department of AutomationShanghai Jiaotong UniversityShanghaiChina

Personalised recommendations