A distance based time series classification framework

Kaya, Huseyin; Gündüz-Ögüdücü, Şule

doi:10.1016/j.is.2015.02.005

Cited by 36 publications

(13 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Given a classification task and a model with parameter values w, the task for a classifier is to output the predicted classŷ based on the input time series, x 0 ðÞ , …,x t ðÞ . The output feature map from the first convolutional layer is then given by convolving each filter w 1 h for h ¼ 1, …,M 1 with the input:…”

Section: Cnn Structurementioning

confidence: 99%

“…Time series classification is widely applied in different fields such as in astronomy [1] to classify the brightness of a target star, in medical science to diagnose cardiac disorders [2] or to recognize human activities [3,4], and in computer science for speech recognition [5,6]. To handle time series classification, several techniques were proposed, which can be aggregated into three categories: model based, distance based and feature based.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

CNN Approaches for Time Series Classification

Sadouk¹

2019

Time Series Analysis - Data, Methods, and Applications

View full text Add to dashboard Cite

Time series classification is an important field in time series data-mining which have covered broad applications so far. Although it has attracted great interests during last decades, it remains a challenging task and falls short of efficiency due to the nature of its data: high dimensionality, large in data size and updating continuously. With the advent of deep learning, new methods have been developed, especially Convolutional Neural Network (CNN) models. In this paper, we present a review of our time series CNN approaches including: (i) a data-level approach based on encoding time series into frequency-domain signals via the Stockwell transform, (ii) an algorithm-level approach based on an adaptive convolutional layer filter that suits the time series in hand, and (iii) another algorithm-level approach adapted to time series classification tasks with limited annotated data, which is a global, fast and lightweight framework based on a transfer learning technique with a source learning task similar or different but related to the target learning task. These approaches are implemented on identifying human activities including normal movements of typical subjects and disorder-related movements such as stereotypical motor movements of autistic subjects. Experimental results show that our approaches improve performance of time series classification.

show abstract

Section: Cnn Structurementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

CNN Approaches for Time Series Classification

Sadouk¹

2019

Time Series Analysis - Data, Methods, and Applications

View full text Add to dashboard Cite

show abstract

“…This approach has been widely applied in time series classification, as it achieves, in conjunction with the DTW distance, the best accuracy ever reached in many benchmark datasets. As such, quite a few studies and reviews include the 1-NN in the time series literature [3][13][16] [31], and, hence, it is not going to be further detailed in this review.…”

Section: A Taxonomy Of Distance Based Time Series Classificationmentioning

confidence: 99%

“…Within the methods employing indefinite kernels, there are different approaches, and for time series classification we have distinguished three main directions (shown in Figure 8). Some of them just learn with the indefinite kernels [31][70] [72][76] [77] using kernel methods that allow this kind of kernels and without taking into consideration that they are indefinite; others argue that the indefiniteness adversely affects the performance and present some alternatives or solutions [20][24] [78]; finally, others focus on a better understanding of these distance kernels in order to investigate the reason for the indefiniteness [27] [79].…”

Section: Indefinite Distance Kernelsmentioning

confidence: 99%

“…However, they obtained good classification results and argued that for the UNIPEN 2 dataset, most of the eigenvalues of the kernel matrix were measured to be non-negative, concluding that somehow, in the given dataset, the proposed kernel matrix is almost PSD. Following the same direction, Jeong et al [77] proposed a variant of GDS DT W which employs the Weighted DTW (WDTW) measure in order to prevent distortions by outliers, while Kaya et al [31] also employed the GDS kernel with SVMs, but instead of using the distance calculated by the DTW, they explored other distances derived from different alignment methods of the series, such as Signal Alignment via Genetic Algorithm (SAGA) [80]. Pree et al [76] proposed a quantitative comparison of different time series similarity measures used either to construct kernels for SVMs or directly for 1-NN classification, concluding that some of the measures benefit from being applied in an SVM, while others do not.…”

Section: Employing Indefinite Kernelsmentioning

confidence: 99%

See 1 more Smart Citation

A review on distance based time series classification

Abanda

Mori

Lozano

2018

Data Min Knowl Disc

231

View full text Add to dashboard Cite

Time series classification is an increasing research topic due to the vast amount of time series data that are being created over a wide variety of fields. The particularity of the data makes it a challenging task and different approaches have been taken, including the distance based approach. 1-NN has been a widely used method within distance based time series classification due to it simplicity but still good performance. However, its supremacy may be attributed to being able to use specific distances for time series within the classification process and not to the classifier itself. With the aim of exploiting these distances within more complex classifiers, new approaches have arisen in the past few years that are competitive or which outperform the 1-NN based approaches. In some cases, these new methods use the distance measure to transform the series into feature vectors, bridging the gap between time series and traditional classifiers. In other cases, the distances are employed to obtain a time series kernel and enable the use of kernel methods for time series classification. One of the main challenges is that a kernel function must be positive semi-definite, a matter that is also addressed within this review. The presented review includes a taxonomy of all those methods that aim to classify time series using a distance based approach, as well as a discussion of the strengths and weaknesses of each method.

show abstract

A Revamp Approach for Training of HMM to Accelerate Classification of 16S rRNA Gene Sequences

Choudhary

Kurhekar

2018

Lecture Notes in Computer Science

View full text Add to dashboard Cite

A distance based time series classification framework

Cited by 36 publications

References 56 publications

CNN Approaches for Time Series Classification

CNN Approaches for Time Series Classification

A review on distance based time series classification

A Revamp Approach for Training of HMM to Accelerate Classification of 16S rRNA Gene Sequences

Contact Info

Product

Resources

About