Abstract—High-dimensional time series data need dimension-reduction strategies to improve the efficiency of computation and indexing. In this paper, we present a dimension-reduction framework for time series. Generally, recent data are much more interesting and significant for predicting future data than old ones. Our basic idea is to reduce to data dimensionality by keeping more detail on recent-pattern data and less detail on older data. We distinguish our work from other recent-biased dimensionreduction techniques by emphasizing on recent-pattern data and not just recent data. We experimentally evaluate our approach with synthetic data as well as real data. Experimental results show that our approach is accurate and effective as it outperforms other well-known techniques.
Index Terms—Time series analysis, dimensionality reduction, data mining.
Cite: Santi Phithakkitnukoon and Carlo Ratti, "A Recent-Pattern Biased Dimension-Reduction Framework for Time Series Data," Journal of Advances in Information Technology, Vol. 1, No. 4, pp. 168-180, November, 2010.doi:10.4304/jait.1.4.168-180
Copyright © 2013-2020. JAIT. All Rights Reserved
This work is licensed under the Creative Commons Attribution License (CC BY-NC-ND 4.0)