An Empirical Evaluation of Multivariate Time Series Classification with
Input Transformation across Different Dimensions
- URL: http://arxiv.org/abs/2210.07713v2
- Date: Wed, 12 Apr 2023 15:15:17 GMT
- Title: An Empirical Evaluation of Multivariate Time Series Classification with
Input Transformation across Different Dimensions
- Authors: Leonardos Pantiskas, Kees Verstoep, Mark Hoogendoorn, Henri Bal
- Abstract summary: We show that the best transformation-dimension configuration leads to an increase in the accuracy compared to the result of each model.
We also show that if we keep the transformation method constant, there is a statistically significant difference in accuracy results when applying it across different dimensions.
- Score: 3.5786621294068373
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: In current research, machine and deep learning solutions for the
classification of temporal data are shifting from single-channel datasets
(univariate) to problems with multiple channels of information (multivariate).
The majority of these works are focused on the method novelty and architecture,
and the format of the input data is often treated implicitly. Particularly,
multivariate datasets are often treated as a stack of univariate time series in
terms of input preprocessing, with scaling methods applied across each channel
separately. In this evaluation, we aim to demonstrate that the additional
channel dimension is far from trivial and different approaches to scaling can
lead to significantly different results in the accuracy of a solution. To that
end, we test seven different data transformation methods on four different
temporal dimensions and study their effect on the classification accuracy of
five recent methods. We show that, for the large majority of tested datasets,
the best transformation-dimension configuration leads to an increase in the
accuracy compared to the result of each model with the same hyperparameters and
no scaling, ranging from 0.16 to 76.79 percentage points. We also show that if
we keep the transformation method constant, there is a statistically
significant difference in accuracy results when applying it across different
dimensions, with accuracy differences ranging from 0.23 to 47.79 percentage
points. Finally, we explore the relation of the transformation methods and
dimensions to the classifiers, and we conclude that there is no prominent
general trend, and the optimal configuration is dataset- and
classifier-specific.
Related papers
- MEMTO: Memory-guided Transformer for Multivariate Time Series Anomaly
Detection [6.16984478518058]
MEMTO is a memory-guided Transformer that learns the degree to which each memory item should be updated in response to the input data.
We evaluate our proposed method on five real-world datasets from diverse domains.
arXiv Detail & Related papers (2023-12-05T06:28:19Z) - Compatible Transformer for Irregularly Sampled Multivariate Time Series [75.79309862085303]
We propose a transformer-based encoder to achieve comprehensive temporal-interaction feature learning for each individual sample.
We conduct extensive experiments on 3 real-world datasets and validate that the proposed CoFormer significantly and consistently outperforms existing methods.
arXiv Detail & Related papers (2023-10-17T06:29:09Z) - Binary Quantification and Dataset Shift: An Experimental Investigation [54.14283123210872]
Quantification is the supervised learning task that consists of training predictors of the class prevalence values of sets of unlabelled data.
The relationship between quantification and other types of dataset shift remains, by and large, unexplored.
We propose a fine-grained taxonomy of types of dataset shift, by establishing protocols for the generation of datasets affected by these types of shift.
arXiv Detail & Related papers (2023-10-06T20:11:27Z) - HFN: Heterogeneous Feature Network for Multivariate Time Series Anomaly
Detection [2.253268952202213]
We propose a novel semi-supervised anomaly detection framework based on a heterogeneous feature network (HFN) for MTS.
We first combine the embedding similarity subgraph generated by sensor embedding and feature value similarity subgraph generated by sensor values to construct a time-series heterogeneous graph.
This approach fuses the state-of-the-art technologies of heterogeneous graph structure learning (HGSL) and representation learning.
arXiv Detail & Related papers (2022-11-01T05:01:34Z) - Predicting Out-of-Domain Generalization with Neighborhood Invariance [59.05399533508682]
We propose a measure of a classifier's output invariance in a local transformation neighborhood.
Our measure is simple to calculate, does not depend on the test point's true label, and can be applied even in out-of-domain (OOD) settings.
In experiments on benchmarks in image classification, sentiment analysis, and natural language inference, we demonstrate a strong and robust correlation between our measure and actual OOD generalization.
arXiv Detail & Related papers (2022-07-05T14:55:16Z) - Deep learning model solves change point detection for multiple change
types [69.77452691994712]
A change points detection aims to catch an abrupt disorder in data distribution.
We propose an approach that works in the multiple-distributions scenario.
arXiv Detail & Related papers (2022-04-15T09:44:21Z) - Missing Value Imputation on Multidimensional Time Series [16.709162372224355]
We present DeepMVI, a deep learning method for missing value imputation in multidimensional time-series datasets.
DeepMVI combines fine-grained and coarse-grained patterns along a time series, and trends from related series across categorical dimensions.
Experiments show that DeepMVI is significantly more accurate, reducing error by more than 50% in more than half the cases.
arXiv Detail & Related papers (2021-03-02T09:55:05Z) - Generalized Matrix Factorization: efficient algorithms for fitting
generalized linear latent variable models to large data arrays [62.997667081978825]
Generalized Linear Latent Variable models (GLLVMs) generalize such factor models to non-Gaussian responses.
Current algorithms for estimating model parameters in GLLVMs require intensive computation and do not scale to large datasets.
We propose a new approach for fitting GLLVMs to high-dimensional datasets, based on approximating the model using penalized quasi-likelihood.
arXiv Detail & Related papers (2020-10-06T04:28:19Z) - Spherical Feature Transform for Deep Metric Learning [58.35971328774927]
This work proposes a novel spherical feature transform approach.
It relaxes the assumption of identical covariance between classes to an assumption of similar covariances of different classes on a hypersphere.
We provide a simple and effective training method, and in depth analysis on the relation between the two different transforms.
arXiv Detail & Related papers (2020-08-04T11:32:23Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.