59,128 Hits in 7.3 sec

Kernel distance measures for time series, random fields and other structured data [article]

Srinjoy Das, Hrushikesh Mhaskar, Alexander Cloninger
2021 arXiv   pre-print
This paper introduces kdiff, a novel kernel-based measure for estimating distances between instances of time series, random fields and other forms of structured data.  ...  Applications are demonstrated for clustering of synthetic and real-life time series and image data, and the performance of kdiff is compared to competing distance measures for clustering.  ...  Acknowledgements This work was supported in part by NSF awards CNS1730158, ACI-1540112, ACI-1541349, OAC-1826967, the University of California Office of the President, and the California Institute for  ... 
arXiv:2109.14752v1 fatcat:wcjkt4ohx5cilmxjywki56kxby

Kernel Distance Measures for Time Series, Random Fields and Other Structured Data

Srinjoy Das, Hrushikesh N. Mhaskar, Alexander Cloninger
2021 Frontiers in Applied Mathematics and Statistics  
This paper introduces kdiff, a novel kernel-based measure for estimating distances between instances of time series, random fields and other forms of structured data.  ...  Applications are demonstrated for clustering of synthetic and real-life time series and image data, and the performance of kdiff is compared to competing distance measures for clustering.  ...  , a novel kernel-based measure for estimating distances between instances of time series, random fields and other forms of structured data.  ... 
doi:10.3389/fams.2021.787455 fatcat:7sauwbxtj5botneuwtlixmkuje


Michal Švanda, Thierry Roudier, Michel Rieutord, Raymond Burston, Laurent Gizon
2013 Astrophysical Journal  
We compare measurements of horizontal flows on the surface of the Sun using helioseismic time--distance inversions and coherent structure tracking of solar granules.  ...  Further, we find a good agreement between these measurements and the time-averaged Doppler line-of-sight velocity, and also perform sanity checks on the vertical flow that resulted from the 3-D time--distance  ...  Measured surface flows We measured surface flow fields using the two distinct methods (tracking and timedistance inversion) applied to data series observed by Helioseismic and Magnetic Imager (HMI; Scherrer  ... 
doi:10.1088/0004-637x/771/1/32 fatcat:csd55xye4fdctmrkbbsy62byka

Similarity Measure Selection for Clustering Time Series Databases using Leading Activities

T. Karthikeyan, Dr. T. Sitamahalakshmi
2017 International Journal of Engineering Research and  
The goal of data mining method is obtaining information from knowledge set and enhancing it to lucid structure for the further use.  ...  It's a laborious task to extract all data from the given set of the time series. This paper concentrates on the comparative study of different papers and proposes a typical solution for the drawbacks.  ...  Distance Measures for Time Series Data The experimental is re-implementing eight different time series representation and nine similarity measures and their variants, 38 time series data sets from a wide  ... 
doi:10.17577/ijertv6is050557 fatcat:stdq4m2qpfcjvnyegd7zrv6nwq

Remote sensing classification method of vegetation dynamics based on time series Landsat image: a case of opencast mining area in China

Jiaxing Xu, Hua Zhao, Pengcheng Yin, Duo Jia, Gang Li
2018 EURASIP Journal on Image and Video Processing  
And the Markov random field optimized (MRF) semi-supervised dynamic time warping (DTW) kernel fuzzy c-means clustering was constructed.  ...  This method uses the time-division algorithm for fitting time-series NDVI firstly.  ...  Acknowledgements The authors would like to thank the editor, an associate editor, and referees for comments and suggestions which greatly improved this paper. Availability of data materials  ... 
doi:10.1186/s13640-018-0360-0 fatcat:lybmf7wjmnbihduhjgo25uovd4

Stochastic Local Interaction Model with Sparse Precision Matrix for Space-Time Interpolation [article]

Dionissios T. Hristopulos, Vasiliki D. Agou
2019 arXiv   pre-print
For data on regular space-time lattices, the \sli model is equivalent to a Gaussian Markov Random Field.  ...  We formulate a stochastic local interaction (\sli) model for regular and scattered space-time data that incorporates interactions within controlled space-time neighborhoods.  ...  Valerie Monbet (Université de Rennes) for suggesting and helping with the acquisition of the ERA5 reanalysis data. Finally, we thank Dr.  ... 
arXiv:1902.07780v1 fatcat:rmoreuzdbzb3zmria52tn34sve

Costs and benefits of non-random seed release for long-distance dispersal in wind-dispersed plant species

Jelle Treep, Monique de Jager, Leandra S. Kuiper, Tomer Duman, Gabriel G. Katul, Merel B. Soons
2018 Oikos  
Bottom row: dispersal kernels for random seed abscission and for the abscission function fitted to the field data.  ...  Top row: dispersal kernels (distance exceedance probabilities) for random seed abscission and for the optimal non-random abscission strategy with no risk included.  ... 
doi:10.1111/oik.04430 fatcat:7kj2toldqnbknkzl4jfxmjivky

Complex Network Construction of Univariate Chaotic Time Series Based on Maximum Mean Discrepancy

2020 Entropy  
The Lorenz system is used to validate the correctness and effectiveness of the proposed method for measuring the similarity.  ...  In this communication, a method of complex network construction is proposed for univariate chaotic time series, which provides a novel way to analyze time series.  ...  Although the proposed method is specific to chaotic time series, it can also be applied to time series in other fields.  ... 
doi:10.3390/e22020142 pmid:33285917 pmcid:PMC7516554 fatcat:aaeipy76xzabvcb2ky27wmpmma

On revolutions

Armand M. Leroi, Ben Lambert, Matthias Mauch, Marina Papadopoulou, Sophia Ananiadou, Staffan I. Lindberg, Patrik Lindenfors
2020 Palgrave Communications  
Our method is general and can be applied to any phenomenon captured by multivariate time series data of sufficient quality.  ...  We define revolutions as those periods of time when the value of this measure is, by a non-parametric test, shown to significantly exceed the background rate.  ...  Acknowledgements We thank Simon deDeo, Tim Hitchcock, and Matthew Jockers for supplying data. Arnold Heumakers, Tim Hitchcock, James Rosindell, and Ted Underwood for comments on the paper.  ... 
doi:10.1057/s41599-019-0371-1 fatcat:gtei63znwrdaxipi6lfpvjl5c4

Damage detection in an uncertain nonlinear beam based on stochastic Volterra series: An experimental application

Luis G.G. Villani, Samuel da Silva, Americo Cunha, Michael D. Todd
2019 Mechanical systems and signal processing  
Mechanical Systems and Signal Processing, Elsevier, 2019, 128, pp.obtained by the deterministic version of the Volterra series, showing the advantage of the stochastic model use when we consider the experimental  ...  data variation with the capability to detect the presence of the damage with statistical confidence.  ...  from a different time series caused by a sine input signal with a different fundamental frequency.  ... 
doi:10.1016/j.ymssp.2019.03.045 fatcat:mtmps7s7v5cxpce4mvnunrmz7a

Deep RBFNet: Point Cloud Feature Learning using Radial Basis Functions [article]

Weikai Chen, Xiaoguang Han, Guanbin Li, Chao Chen, Jun Xing, Yajie Zhao, Hao Li
2019 arXiv   pre-print
A typical RBF kernel, e.g. Gaussian function, naturally penalizes long-distance response and is only activated by neighboring points.  ...  In addition, our framework allows the joint optimization of kernel distribution and its receptive field, automatically evolving kernel configurations in an end-to-end manner.  ...  The distance metric (Euclidean distance) for measuring spatial point distribution is particularly suitable for RBF kernel to encode and learn from.  ... 
arXiv:1812.04302v2 fatcat:ma3vjx47mbcsnjoje3tljrdpq4

A Distance Covariance-based Kernel for Nonlinear Causal Clustering in Heterogeneous Populations [article]

Alex Markham and Richeek Das and Moritz Grosse-Wentrup
2022 arXiv   pre-print
To this end, we introduce a distance covariance-based kernel designed specifically to measure the similarity between the underlying nonlinear causal structures of different samples.  ...  Though we focus on the theoretical aspects of the kernel, we also evaluate its performance on synthetic data and demonstrate its use on a real gene expression data set.  ...  Acknowledgments We thank Anja Meunier (University of Vienna) and Liam Solus (KTH Royal Institute of Technology) for helpful discussions and comments on a previous draft.  ... 
arXiv:2106.03480v2 fatcat:qbmqflgw6zailluwr3nkbh4g3a

A review on distance based time series classification

Amaia Abanda, Usue Mori, Jose A. Lozano
2018 Data mining and knowledge discovery  
In other cases, the distances are employed to obtain a time series kernel and enable the use of kernel methods for time series classification.  ...  Time series classification is an increasing research topic due to the vast amount of time series data that is being created over a wide variety of fields.  ...  Acknowledgements The authors would like to thank the people who contributed to the UCR time series repository, as well as would like to express our sincere appreciation for the comments and advices provided  ... 
doi:10.1007/s10618-018-0596-4 fatcat:t57qk2xy5vfutj6scb5zc2fnoa


Michal Švanda
2013 Astrophysical Journal  
k--\omega filtering procedures -- ridge filtering and phase-speed filtering -- commonly used in time--distance helioseismology.  ...  The consistency of time--distance inversions for horizontal components of the plasma flow on supergranular scales in the upper solar convection zone is checked by comparing the results derived using two  ...  of wave travel times for distances shorter than 10 Mm.  ... 
doi:10.1088/0004-637x/775/1/7 fatcat:mhkoyv6pqzdwdf37vclp42hvbq

Comparison of Machine Learning Approaches to Improve Diagnosis of Optic Neuropathy Using Photopic Negative Response Measured Using a Handheld Device

Tina Diao, Fareshta Kushzad, Megh D. Patel, Megha P. Bindiganavale, Munam Wasi, Mykel J. Kochenderfer, Heather E. Moss
2021 Frontiers in Medicine  
However, technical requirements for measurement and poor classification performance have limited widespread clinical application.  ...  In comparison, machine learning classifiers using a variety of time series analysis approaches had F1 scores of 0.58–0.76 on a test data set.  ...  The KBF kernel for a pair of data series x 1 and another data series x 2 is exp (-γ||x 1x 2 || 2 ), where γ is a hyperparameter and ||x 1x 2 || is the Euclidean distance.  ... 
doi:10.3389/fmed.2021.771713 pmid:34926514 pmcid:PMC8677942 fatcat:zdgu6ccgfjgoblztod6oowd4se
« Previous Showing results 1 — 15 out of 59,128 results