Random Dilated Shapelet Transform: A New Approach for Time Series Shapelets

28 Sep 2021  ·  Antoine Guillaume, Christel Vrain, Elloumi Wael ·

Shapelet-based algorithms are widely used for time series classification because of their ease of interpretation, but they are currently outperformed by recent state-of-the-art approaches. We present a new formulation of time series shapelets including the notion of dilation, and we introduce a new shapelet feature to enhance their discriminative power for classification. Experiments performed on 112 datasets show that our method improves on the state-of-the-art shapelet algorithm, and achieves comparable accuracy to recent state-of-the-art approaches, without sacrificing neither scalability, nor interpretability.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Time Series Classification ACSF1 R_DST_Ensemble Accuracy(30-fold) 0.8433333333333333 # 1
Time Series Classification Adiac R_DST_Ensemble Accuracy(30-fold) 0.80230179028133 # 1
Time Series Classification ArrowHead R_DST_Ensemble Accuracy(30-fold) 0.8912380952380949 # 1
Time Series Classification Beef R_DST_Ensemble Accuracy(30-fold) 0.7511111111111111 # 1
Time Series Classification Earthquakes R_DST_Ensemble Accuracy(30-fold) 0.7390887290167865 # 1
Time Series Classification ECG200 R_DST_Ensemble Accuracy(30-fold) 0.9016666666666667 # 1
Time Series Classification ECG5000 R_DST_Ensemble Accuracy(30-fold) 0.9467629629629628 # 1
Time Series Classification Wafer R_DST_Ensemble Accuracy 0.9999513303049968 # 1
Accuracy(30-fold) 0.9999513303049968 # 1


No methods listed for this paper. Add relevant methods here