Use Short Isometric Shapelets to Accelerate Binary Time Series Classification

TL;DR

This paper introduces the short isometric shapelet transform (SIST), a novel method that accelerates binary time series classification by reducing shapelet candidate generation and simplifying classifier training, maintaining accuracy.

Contribution

The paper proposes SIST, a new algorithm that reduces time complexity in shapelet-based classification through fixed shapelet length and single linear classifier training, with theoretical and empirical validation.

Findings

SIST significantly speeds up shapelet transform algorithms.

SIST maintains near-lossless accuracy compared to traditional methods.

Empirical results show superior performance of SIST in classification tasks.

Abstract

In the research area of time series classification, the ensemble shapelet transform algorithm is one of state-of-the-art algorithms for classification. However, its high time complexity is an issue to hinder its application since its base classifier shapelet transform includes a high time complexity of a distance calculation and shapelet selection. Therefore, in this paper we introduce a novel algorithm, i.e. short isometric shapelet transform, which contains two strategies to reduce the time complexity. The first strategy of SIST fixes the length of shapelet based on a simplified distance calculation, which largely reduces the number of shapelet candidates as well as speeds up the distance calculation in the ensemble shapelet transform algorithm. The second strategy is to train a single linear classifier in the feature space instead of an ensemble classifier. The theoretical evidences of these two strategies are presented to guarantee a near-lossless accuracy under some preconditions while reducing the time complexity. Furthermore, empirical experiments demonstrate the superior performance of the proposed algorithm.