The great multivariate time series classification bake off: a review and experimental evaluation of recent algorithmic advances
From MaRDI portal
Publication:2659282
DOI10.1007/S10618-020-00727-3zbMATH Open1478.62267arXiv2007.13156OpenAlexW3115948762WikidataQ112691960 ScholiaQ112691960MaRDI QIDQ2659282FDOQ2659282
Authors: Yanyan Li 
Publication date: 26 March 2021
Published in: Data Mining and Knowledge Discovery (Search for Journal in Brave)
Abstract: Time Series Classification (TSC) involved building predictive models for a discrete target variable from ordered, real valued, attributes. Over recent years, a new set of TSC algorithms have been developed which have made significant improvement over the previous state of the art. The main focus has been on univariate TSC, i.e. the problem where each case has a single series and a class label. In reality, it is more common to encounter multivariate TSC (MTSC) problems where multiple series are associated with a single label. Despite this, much less consideration has been given to MTSC than the univariate case. The UEA archive of 30 MTSC problems released in 2018 has made comparison of algorithms easier. We review recently proposed bespoke MTSC algorithms based on deep learning, shapelets and bag of words approaches. The simplest approach to MTSC is to ensemble univariate classifiers over the multivariate dimensions. We compare the bespoke algorithms to these dimension independent approaches on the 26 of the 30 MTSC archive problems where the data are all of equal length. We demonstrate that the independent ensemble of HIVE-COTE classifiers is the most accurate, but that, unlike with univariate classification, dynamic time warping is still competitive at MTSC.
Full work available at URL: https://arxiv.org/abs/2007.13156
Recommendations
- Deep learning for time series classification: a review
- ROCKET: exceptionally fast and accurate time series classification using random convolutional kernels
- Learning a symbolic representation for multivariate time series classification
- Scalable classifier-agnostic channel selection for multivariate time series classification
- TS-CHIEF: a scalable and accurate forest algorithm for time series classification
Computational methods for problems pertaining to statistics (62-08) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Statistical aspects of big data and data science (62R07) Research exposition (monographs, survey articles) pertaining to statistics (62-02)
Cites Work
- ROCKET: exceptionally fast and accurate time series classification using random convolutional kernels
- Random forests
- Should we really use post-hoc tests based on mean-ranks?
- Statistical comparisons of classifiers over multiple data sets
- Time series shapelets: a novel technique that allows accurate, interpretable and fast classification
- Classification of time series by shapelet transformation
- A time series forest for classification and feature extraction
- Deep learning for time series classification: a review
- Time series representation and similarity based on local autopatterns
- Generalized random shapelet forests
- Interpretable time series classification using linear models and multi-resolution multi-domain symbolic representations
- A probabilistic classifier ensemble weighting scheme based on cross-validated accuracy estimates
Cited In (21)
- A cycle deep belief network model for multivariate time series classification
- Multivariate time-series classification with hierarchical variational graph pooling
- Observer-based dynamical pattern recognition via deterministic learning
- Combining discrete SVM and fixed cardinality warping distances for multivariate time series classification
- Time series classification by Euclidean distance-based visibility graph
- A deep multi-task representation learning method for time series classification and retrieval
- Bake off redux: a review and experimental evaluation of recent time series classification algorithms
- Functional classwise principal component analysis: a classification framework for functional data analysis
- Time series extrinsic regression. Predicting numeric values from time series data
- Random Projection Ensemble Classification with High-Dimensional Time Series
- Deep learning for time series classification: a review
- Scalable classifier-agnostic channel selection for multivariate time series classification
- Neural-symbolic temporal decision trees for multivariate time series classification
- MultiETSC: automated machine learning for early time series classification
- HIVE-COTE 2.0: a new meta ensemble for time series classification
- A novel and effective method for characterizing time series correlations based on martingale difference correlation
- Parameter inference from a non-stationary unknown process
- Co-eye: a multi-resolution ensemble classifier for symbolically approximated time series
- Time series classification based on AdaBoost and deep residual network
- TS-CHIEF: a scalable and accurate forest algorithm for time series classification
- Learning a symbolic representation for multivariate time series classification
Uses Software
This page was built for publication: The great multivariate time series classification bake off: a review and experimental evaluation of recent algorithmic advances
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2659282)
