EFFICIENT LEARNING OF TIME SERIES SHAPELETS

Previously, shapelet generating approaches extracted shapelets from training time series or learned shapelets with many parameters. We first discover shapelet candidates from the Piecewise Aggregate Approximation PAA word space, which is much more efficient than searching in the raw time series space. Distributed optimization and statistical learning via the alternating direction method of multipliers. Extensive experimentation on 15 datasets demonstrates that our algorithm is more accurate against 6 baselines and outperforms 2 orders of magnitude in terms of efficiency. Showing of 8 citations. Skip to search form Skip to main content.

Kwok , Jacek M. Barnaghi , Antonio F. Showing of 8 citations. Existing methods perform a combinatorial search for shapelet discovery. Previously, shapelet generating approaches extracted shapelets from training time series or learned shapelets with many parameters. Articles by Zicheng Fang. Although they can achieve higher accuracy than other approaches, they still confront some challenges.

Moreover, our algorithm has fewer redundant shape-like shapelets and is more convenient to interpret classification decisions.

Recently, time series classification with shapelets, due to their high discriminative ability and good interpretability, has attracted considerable interests within the research community. Boosting the kernelized shapelets: Foundations and Trends in Machine Learning 3 1: By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy EfcicientEfficient of Serviceand Dataset License.

  VADALVAAT EPISODE 6

Skip to search form Skip to main content. Articles by Peng Wang. From This Paper Figures, tables, and topics from this paper.

There was a problem providing the content you requested

Extensive experimentation on 15 datasets demonstrates that our algorithm is more accurate against 6 baselines and outperforms 2 orders of efficinet in terms of efficiency.

Although they can achieve higher accuracy than other approaches, they still confront some challenges.

We first discover shapelet candidates from the Piecewise Aggregate Approximation PAA word space, which is much more efficient than searching in the raw time series space. Zurada Published in AAAI In timeseries classification, shapelets are subsequences of timeseries with high discriminative power. Moreover, the concept of coverage is proposed to measure the quality of candidates, based on which we design a method to compute the optimal efficidnt of shapelets.

References Publications referenced by this paper. First, searching or egficient shapelets in the raw time series space incurs a huge computation cost. Showing of 8 citations.

BarnaghiAntonio F. For example, it may cost several hours to deal with only hundreds of time series.

Efficient Learning of Timeseries Shapelets – Semantic Scholar

Articles by Zicheng Fang. Citations Publications citing this paper. Distributed optimization and statistical learning via the alternating direction method of multipliers. Showing of 25 references. In this paper, we take an entirely different approach and reformulate the shapelet discovery task as a numerical optimization problem.

  MOVIE THEATER 34668

Efficient Learning of Timeseries Shapelets

After that, we apply the logistic regression classifier to adjust the shapelets. Bagnall Data Mining and Knowledge Discovery Previously, shapelet generating leearning extracted shapelets from training time series or learned shapelets with many parameters. Articles by Wei Wang. Even with speedup heuristics such as pruning, clustering, and dimensionality reduction, the search remains computationally expensive.

Second, they must determine how many shapelets are needed beforehand, which is difficult without prior knowledge. Existing methods perform a combinatorial search for shapelet discovery.

KwokJacek M. To overcome these challenges, in this paper, we propose a novel algorithm to learn shapelets. Kwok and Jacek M.