Shapelet Ensemble for Multi-dimensional Time Series

Abstract

Time series shapelets are small subsequences that maximally differentiate classes of time series. Since the inception of shapelets, researchers have used shapelets for various data domains including anthropology and health care, and in the process suggested many efficient techniques for shapelet discovery. However, multi-dimensional time series data poses unique challenges to shapelet discovery that are yet to be solved.We show that an ensemble of shapelet-based decision trees on individual dimensions works better than shapelets defined over multiple dimensions. Generating a shapelet ensemble for multidimensional time series is computationally expensive. Most of the existing techniques prune shapelet candidates for speed. In this paper, we propose a novel technique for shapelet discovery that evaluates remaining candidates efficiently. Our algorithm uses a multi-length approximate index for time series data to efficiently find the nearest neighbors of the candidate shapelets. We employ a simple skipping technique for additional candidate pruning and a voting based technique to improve accuracy while retaining interpretability. Not only do we find a significant speed increase, our techniques enable us to efficiently discover shapelets on datasets with multi-dimensional and long time series such as hours of brain activity recordings. We demonstrate our approach on a biomedical dataset and find significant differences between patients with schizophrenia and healthy controls.

Sensor-based human activity recognition system with a multilayered model using time series shapelets

“…The second solution is to generate an ensemble of shapelet-1060 based decision trees for complex activity recognition. Cetin et al [76] proposed a novel ensembling technique for finding the nearest neighbors of the candidate shapelets efficiently. They claimed that their technique can efficiently discover shapelets on datasets with multi-dimensional and long time series, which 1065 will be useful in our activity recognition scenarios involving multi-source sensed data (e.g.…”

“…Many optimization techniques [75, 67,76] can provide a near-optimal solution in reasonable time, including metaheuristic algorithms or nature inspired artificial techniques. Re-795 gardless of the method used for training shapelets, they are computed on the server.…”

Locally and globally explainable time series tweaking

“…The algorithm employs the Fast Shapelet selection approach for extracting the most informative shapelets per dimension. In a similar manner, a shapelet tree is built from each time series dimension using several additional techniques for providing search speedups [5]. Moreover, various voting approaches are evaluated for providing the final classification label, demonstrating that one shapelet tree per dimension outperforms shapelets defined over multiple dimensions [5].…”

“…In a similar manner, a shapelet tree is built from each time series dimension using several additional techniques for providing search speedups [5]. Moreover, various voting approaches are evaluated for providing the final classification label, demonstrating that one shapelet tree per dimension outperforms shapelets defined over multiple dimensions [5]. More recently, the generalized random shapelet forest has been proposed for univariate and multivariate time series classification, by expanding the idea of random shapelet trees and randomly selecting shapelet features per dimension [20].…”

Data Mining with Shapelets for Predicting Valve Failures in Gas Compressors

“…There are a few approaches (such as [2,3,8]), which propose extensions of the univariate shapelet extraction method for multivariate use cases, but all of them make the assumption that shapelets can be extracted from different sensors independently of each other. This assumption does not hold for real-world oilfield data where one sensor may affect the reading of other sensors (e.g.…”