Generalized Dynamic Time Warping: Unleashing the Warping Power Hidden in Point-Wise Distances

Neamtu, Rodica; Ahsan, Ramoza; Rundensteiner, Elke A.; Sárközy, Gábor N.; Keogh, Eamonn; Dau, Son Hoang; Nguyen, Cuong; Lovering, Charles

doi:10.1109/icde.2018.00054

Cited by 17 publications

(26 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…And finally: Could the black-box effect of DNNs be avoided to provide interpretability? Given that the latter questions have not been addressed by the TSC community, it is surprising how much recent papers have neglected the possibility that TSC problems could be solved using a pure feature learning algorithm (Neamtu et al, 2018;Bagnall et al, 2017;Lines et al, 2016). In fact, a recent empirical study (Bagnall et al, 2017) evaluated 18 TSC algorithms on 85 time series datasets, none of which was a deep learning model.…”

Section: Introductionmentioning

confidence: 99%

Deep learning for time series classification: a review

Fawaz

Forestier

Weber

et al. 2019

Data Min Knowl Disc

2,070

712

View full text Add to dashboard Cite

Time Series Classification (TSC) is an important and challenging problem in data mining. With the increase of time series data availability, hundreds of TSC algorithms have been proposed. Among these methods, only a few have considered Deep Neural Networks (DNNs) to perform this task. This is surprising as deep learning has seen very successful applications in the last years. DNNs have indeed revolutionized the field of computer vision especially with the advent of novel deeper architectures such as Residual and Convolutional Neural Networks. Apart from images, sequential data such as text and audio can also be processed with DNNs to reach state-of-the-art performance for document classification and speech recognition. In this article, we study the current state-ofthe-art performance of deep learning algorithms for TSC by presenting an empirical study of the most recent DNN architectures for TSC. We give an overview of the most successful deep learning applications in various time series domains under a unified taxonomy of DNNs for TSC. We also provide an open source deep learning framework to the TSC community where we implemented each of the compared approaches and evaluated them on a univariate TSC benchmark (the UCR/UEA archive) and 12 multivariate time series datasets. By training 8,730 deep learning models on 97 time series datasets, we propose the most exhaustive study of DNNs for TSC to date.

show abstract

Section: Introductionmentioning

confidence: 99%

Deep learning for time series classification: a review

Fawaz

Forestier

Weber

et al. 2019

Data Min Knowl Disc

2,070

712

View full text Add to dashboard Cite

show abstract

“…Basically it's a pairwise comparison of the feature vectors of time series data.The classification is based on the nearest match of that particular sample with the test sample through the distance measure technique. DTW is effective in this case as there may have missing information or varying lengths of the vectors, on condition that the sequences are long enough for matching [11]. The above figure (Fig.…”

Section: Dynamic Time Warpingmentioning

confidence: 99%

Development of a System with Online Signature Biometric for Access Control Application

Hazra¹,

Chaudhuri²,

Bandyopadhyay³

2019

dtcse

View full text Add to dashboard Cite

Online signature verification system is a technique based on behavioural biometrics which is gaining popularity for its widespread acceptability due to ease of use and forge proof features. Online (dynamic) signatures are captured from pressure sensitive tablets, signature pads, tablet PCs for further processing. This paper depicts the working mechanism of such a system capable of verifying the true signers as well as detecting the forgery attempt. After the signature data collection, the signatures were preprocessed for future use. A set of total 58 different features (local, global and normalized) were extracted from the online signature data. The system presents a binary classification technique since the test signature is categorized into one of the two classes (genuine or false). Template matching based Dynamic Time Warping (DTW) classifier is used for training and testing purposes with an average recognition rate of 90% for true signers. A separate version of the system was also developed which blocks random and skilled forgeries with a high level of accuracy though the TN rate is higher than the system described here.

show abstract

“…Further improvements to this approach were made by using other types of classifiers (Bagnall et al, 2016;Baydogan et al, 2013;Bostrom & Bagnall, 2015;Deng et al, 2013) which were found to be more effective than NN-DTW (Bagnall et al, 2017). Nevertheless, Fawaz et al (2019) criticized the community for not using deep neural networks (DNNs) in comparison studies of ensemble methods (Bagnall et al, 2017;Lines et al, 2016;Neamtu et al, 2018).…”

Section: Classification With Time Seriesmentioning

confidence: 99%

Multivariate temporal data analysis ‐ a review

Moskovitch

2021

WIREs Data Min & Knowl

View full text Add to dashboard Cite

The information technology revolution, especially with the adoption of the Internet of Things, longitudinal data in many domains become more available and accessible for secondary analysis. Such data provide meaningful opportunities to understand process in many domains along time, but also challenges. A main challenge is the heterogeneity of the temporal variables due to the different types of data, whether a measurement or an event, and type of samplings: fixed or irregular. Other variables can be also events that may or not have duration. In this review, we discuss the various types of temporal data, and the various relevant analysis methods. Starting with fixed frequency variables, with forecasting and time series methods, and proceeding with sequential data, and sequential patterns mining, and time intervals mining for events having various time duration. Also the use of various deep learning based architectures for temporal data is discussed. The challenge of heterogeneous multivariate temporal data analysis and discuss various options to deal with it, focusing on an increasingly used option of transforming the data into symbolic time intervals through temporal abstraction and the use of time intervals related patterns discovery for temporal knowledge discovery, clustering, classification prediction, and more. Finally, we discuss the overview of the field, and areas in which more studies and contributions are needed. This article is categorized under: Algorithmic Development > Spatial and Temporal Data Mininganalytics, temporal data mining, time series | INTRODUCTIONWith the introduction of the internet, an exponential growth in text based data, which required new methods to analyze text, retrieve, understand, classify, and more. However, with the recent growing adoption of the Internet of Things, in which devices in various domains become connected to the internet network and produce multitude of logged longitudinal data, we are facing a similar increase in multivariate temporal data that is generated by various devices and sensors. Such data can be logged from smart watches, cars, smart refrigerators, smart phones, televisions, sensors in the space of rooms for various purposes, and many more, which can be useful in various domains, and for different tasks. This trend is expected to increase significantly and requires new methods and more scalability.An extension of temporal data analytics, which are out of the scope of this article, is spatio-temporal data analysis (Gong

show abstract

Generalized Dynamic Time Warping: Unleashing the Warping Power Hidden in Point-Wise Distances

Cited by 17 publications

References 29 publications

Deep learning for time series classification: a review

Deep learning for time series classification: a review

Development of a System with Online Signature Biometric for Access Control Application

Multivariate temporal data analysis ‐ a review

Contact Info

Product

Resources

About