Dimensionality Reduction by Turning Points for Stream Time Series Prediction
- 1.1k Downloads
In recent years, there has been an explosion of problems concerned with mining time series databases and dimensionality reduction is one of the important tasks in time series data mining analysis. These approaches are very useful to pre-process the large dataset and then use it to analyze and mine. In this paper, we propose a method based on turning points to reduce the dimensions of stream time series data, this task helps the prediction process faster. The turning points which are extracted from the maximum or minimum points of the time series stream are proved more efficient and effective in preprocessing data for stream time series prediction. To implement the proposed framework, we use stock time series obtained from Yahoo Finance, the prediction approach based on Sequential Minimal Optimization and the experimental results validate the effectiveness of our approach.
KeywordsStream Mining Time Series Dimensionality Reduction Turning Points Stream Time Series Time Series Prediction
Unable to display preview. Download preview PDF.
- 5.Zhang, Z., Li, J., Wang, H., Wang, S.: Study of principal component analysis on multi-dimension stock data. Chinese Journal of Scientific Instrument 26(8), 2489–2491 (2005)Google Scholar
- 9.Yang, J.F., Zhai, Y.J., Xu, D.P., Han, P.: SMO algorithm applied in time series model building and forecast. In: Proc. of the 6th International Conference on Machine Learning and Cybernetics, Hong Kong, vol. 4, pp. 2395–2400 (2007)Google Scholar
- 10.Osuna, E., Freund, R.: An improved training algorithm for support vector machines. In: Proceedings of the 1997 IEEE Neural Networks for Signal Processing VII, New York, pp. 276–285. IEEE (1997)Google Scholar
- 11.Platt, J.C.: Sequential Minimal Optimization: A fast algorithm for training support vector machines. Advances in kernel methods, pp. 185–208. MIT Press, Cambridge (1999)Google Scholar
- 12.Chen, Z., Yang, Y.: Assessing Forecast Accuracy Measures (2004), http://www.stat.iastate.edu/preprint/articles/2004-10.pdf
- 13.Keogh, E., Chakrabarti, K., Pazzani, M., Mehrotra, S.: Locally Adaptive Dimensionality Reduction for Indexing Large Time Series Databases. In: Proceedings of ACM SIGMOD Conference on Management of Data, pp. 151–162 (2001)Google Scholar