Two Approaches to Machine Learning Classification of Time Series Based on Recurrence Plots

Kirichenko, Lyudmyla; Радівілова, Тамара; Bulakh, Vitalii; Zinchenko, Petro; Alghawli, Abed Saif

doi:10.1109/dsmp47368.2020.9204021

Cited by 12 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, an RM space method is adopted to diagnose such faults. [34] For a sampled single variable x, assuming that the time window length is T, and the sampled value is recorded as…”

Section: Rm Spacementioning

confidence: 99%

Multi‐block gated recurrent unit network‐based fault diagnosis method for multivariate nonlinear process

et al. 2023

View full text Add to dashboard Cite

The increasing complexity of industrial processes brings new challenges to fault diagnosis tasks, and different types of faults have higher and higher requirements on the performance of fault classification models. This paper proposes a novel multivariate nonlinear temporal‐related fault diagnosis method based on gated recurrent units (GRUs). First, to improve the performance of the model in local information extraction and global integration, high dimensional variables are divided into multiple sub‐blocks according to the structure of chemical process units, and a new block normalization method is proposed to improve the performance of local feature extraction. Second, aiming at the slow drifting faults, the GRU network is adopted inside the sub‐block to extract local sparse and nonlinear temporal features. By combining the variance features of variables after block normalization, the performance of the model on multiplicative faults will improve. Finally, aiming at the complex correlation between variables, a new recurrent matrix method is proposed to extract the time transform information inside each variable to improve the comprehensive performance of the model. Through a multi‐level feature integration strategy, the model can be trained in parallel to improve the training speed. The proposed method shows good performance in the Tennessee Eastman process, and the extracted multi‐class features allow the model to be trained end‐to‐end and simultaneously diagnose multiple types of faults.

show abstract

“…Therefore, an RM space method is adopted to diagnose such faults. [34] For a sampled single variable x, assuming that the time window length is T, and the sampled value is recorded as…”

Section: Rm Spacementioning

confidence: 99%

Multi‐block gated recurrent unit network‐based fault diagnosis method for multivariate nonlinear process

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Another group of works discussed methods for assessing the competence of experts [12,13]. The problem of finding an optimal solution is approached as a «game with nature», where a number of methods and criteria were developed [14,15]. The choice of criteria for making decisions usually depends on the decision maker, analysts and the owner of the problem, and depends on such subjective characteristics as the level of optimism, risk tolerance, etc.…”

Section: Research Of Existing Solution To the Problemmentioning

confidence: 99%

Development of decision approval rules in multichannel decision-making systems

Mulesa

Bilak

Kykyna

et al. 2021

TAPR

View full text Add to dashboard Cite

The research is devoted to the development of rules for the coordination of decisions in multichannel decision-making systems. Systems are considered that in an automated continuous mode process incoming signals from different channels and, on their basis, make the final decision. One of the most problematic stages in the operation of such systems is their own coordination of solutions received from different channels. There may be cases where different channels provide signals with opposite values. Then the choice of the decisive solution should depend on the reliability of the channels under consideration. The object of research is the processes that take place during the coordination of decisions in multichannel decision-making systems. The development and implementation of such systems will allow in an automated mode to generalize the solution obtained through different channels, to increase the reliability and efficiency of the systems as a whole. During the study, the following methods were used: – a systematic approach – when analyzing the structure and functioning of multichannel one-stage decision-making systems; – method of mathematical modeling – for formalizing the problem of coordinating decisions in multichannel decision-making systems; – method of analysis – when developing rules for agreeing decisions. The authors analyzed the structure of a one-stage multichannel decision-making system. The case is considered when the channels, based on the initial data entering the system, decide on the presence or absence of a certain fact. That is, the channels send signals from the set {True, False}. In the study, decision rules for the coordination of decisions were developed, taking into account not only the signals received from different channels, but also the reliability of the channels themselves. As is usual in decision theory, different rules can give different results for the same initial data. The choice of the decision rule depends on the decision maker, its personal psychological qualities and the scope of the system.

show abstract

“…Recurrence plots are also used to create 2D images from sensors 1D signals. There, distance matrices capture temporal patterns in the signal and represent it as texture patterns in the image [ 22 , 23 , 24 ]. Another approach, is to construct an image using Fourier Transformation and create a spectrogram [ 25 , 26 ].…”

Section: Introductionmentioning

confidence: 99%

INIM: Inertial Images Construction with Applications to Activity Recognition

Daniel

Klein

2021

Sensors

View full text Add to dashboard Cite

Human activity recognition aims to classify the user activity in various applications like healthcare, gesture recognition and indoor navigation. In the latter, smartphone location recognition is gaining more attention as it enhances indoor positioning accuracy. Commonly the smartphone’s inertial sensor readings are used as input to a machine learning algorithm which performs the classification. There are several approaches to tackle such a task: feature based approaches, one dimensional deep learning algorithms, and two dimensional deep learning architectures. When using deep learning approaches, feature engineering is redundant. In addition, while utilizing two-dimensional deep learning approaches enables to utilize methods from the well-established computer vision domain. In this paper, a framework for smartphone location and human activity recognition, based on the smartphone’s inertial sensors, is proposed. The contributions of this work are a novel time series encoding approach, from inertial signals to inertial images, and transfer learning from computer vision domain to the inertial sensors classification problem. Four different datasets are employed to show the benefits of using the proposed approach. In addition, as the proposed framework performs classification on inertial sensors readings, it can be applied for other classification tasks using inertial data. It can also be adopted to handle other types of sensory data collected for a classification task.

show abstract

Two Approaches to Machine Learning Classification of Time Series Based on Recurrence Plots

Cited by 12 publications

References 24 publications

Multi‐block gated recurrent unit network‐based fault diagnosis method for multivariate nonlinear process

Multi‐block gated recurrent unit network‐based fault diagnosis method for multivariate nonlinear process

Development of decision approval rules in multichannel decision-making systems

INIM: Inertial Images Construction with Applications to Activity Recognition

Contact Info

Product

Resources

About