Deep attentive time warping

Matsuo, Shinnosuke; Wu, Xiaomeng; Atarsaikhan, Gantugs; Kimura, Atsushi; Kashino, Kunio; Iwana, Brian Kenji; Uchida, Seiichi

doi:10.1016/j.patcog.2022.109201

Cited by 3 publications

(2 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The following section explores the existing literature on time-series classification. This body of work is vast and varied, encompassing a range of methodologies from time-warping [14][15][16] and feature-based methods [17][18][19][20] to ensemble-based methods [21,22], deep neural networks, and hybrid methods [23][24][25]. Each approach produces unique insights and challenges, contributing to the rich tapestry of research.…”

Section: Related Workmentioning

confidence: 99%

Wide-TSNet: A Novel Hybrid Approach for Bitcoin Price Movement Classification

Yamak,

Li,

Zhang

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

In this paper, we introduce Wide-TSNet, a novel hybrid approach for predicting Bitcoin prices using time-series data transformed into images. The method involves converting time-series data into Markov transition fields (MTFs), enhancing them using histogram equalization, and classifying them using Wide ResNets, a type of convolutional neural network (CNN). We propose a tripartite classification system to accurately represent Bitcoin price trends. In addition, we demonstrate the effectiveness of Wide-TSNet through various experiments, in which it achieves an Accuracy of approximately 94% and an F1 score of 90%. It is also shown that lightweight CNN models, such as SqueezeNet and EfficientNet, can be as effective as complex models under certain conditions. Furthermore, we investigate the efficacy of other image transformation methods, such as Gramian angular fields, in capturing the trends and volatility of Bitcoin prices and revealing patterns that are not visible in the raw data. Moreover, we assess the effect of image resolution on model performance, emphasizing the importance of this factor in image-based time-series classification. Our findings explore the intersection between finance, image processing, and deep learning, providing a robust methodology for financial time-series classification.

show abstract

Section: Related Workmentioning

confidence: 99%

Wide-TSNet: A Novel Hybrid Approach for Bitcoin Price Movement Classification

Yamak,

Li,

Zhang

et al. 2024

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Siamese networks have been used in a wide range of contexts such as visual tracking [29,30,31], signature recognition [32,33,34], anomaly detection [35,36,37] and speech signal processing [38,39,40]. This technique has shown an excellent performance handling with heterogeneous data [41,42,43], including the evaluation of medical images [44,45,46].…”

Section: Related Workmentioning

confidence: 99%

Ensembling shallow siamese architectures to assess functional asymmetry in Alzheimer’s disease progression

Arco

Ortíz²,

Castillo-Barnes³

et al. 2023

Applied Soft Computing

View full text Add to dashboard Cite

Research on the design of a novel composite solenoid model ship simulation magnetic source based on DTW

Lu,

Zhang

2024

AIP Advances

View full text Add to dashboard Cite

In order to study the simulation ability of a solenoid on a ship magnetic field, a new composite model based on a double row magnetic dipole array model with an ellipsoid was designed. The time-domain characteristics, such as magnetic field characteristics, magnetic inclination cosine, magnetic field scalar gradient, and magnetic field spatial distribution characteristics, were extracted. It was verified that the model can accurately simulate the magnetic field characteristics of ships through ship model experiments and simulation analysis. To further quantitatively and accurately analyze the simulation ability of the composite model for ship magnetic fields, a dynamic time wrapping measurement method based on one-dimensional time series was adopted to calculate the similarity of four characteristic quantities: magnetic field vector, magnetic field modulus, magnetic inclination cosine, and magnetic field scalar gradient. The results showed that the composite model has high simulation ability for ship magnetic fields.

show abstract

Deep attentive time warping

Cited by 3 publications

References 39 publications

Wide-TSNet: A Novel Hybrid Approach for Bitcoin Price Movement Classification

Wide-TSNet: A Novel Hybrid Approach for Bitcoin Price Movement Classification

Ensembling shallow siamese architectures to assess functional asymmetry in Alzheimer’s disease progression

Research on the design of a novel composite solenoid model ship simulation magnetic source based on DTW

Contact Info

Product

Resources

About