{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T18:10:23Z","timestamp":1775067023987,"version":"3.50.1"},"reference-count":44,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2022,7,31]],"date-time":"2022-07-31T00:00:00Z","timestamp":1659225600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Ministry of Research, Innovation and Digitization, CNCS-UEFISCDI","award":["PN-III-P4-PCE-2021-0154"],"award-info":[{"award-number":["PN-III-P4-PCE-2021-0154"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>The problem addressed by dictionary learning (DL) is the representation of data as a sparse linear combination of columns of a matrix called dictionary. Both the dictionary and the sparse representations are learned from the data. We show how DL can be employed in the imputation of multivariate time series. We use a structured dictionary, which is comprised of one block for each time series and a common block for all the time series. The size of each block and the sparsity level of the representation are selected by using information theoretic criteria. The objective function used in learning is designed to minimize either the sum of the squared errors or the sum of the magnitudes of the errors. We propose dimensionality reduction techniques for the case of high-dimensional time series. For demonstrating how the new algorithms can be used in practical applications, we conduct a large set of experiments on five real-life data sets. The missing data (MD) are simulated according to various scenarios where both the percentage of MD and the length of the sequences of MD are considered. This allows us to identify the situations in which the novel DL-based methods are superior to the existing methods.<\/jats:p>","DOI":"10.3390\/e24081057","type":"journal-article","created":{"date-parts":[[2022,7,31]],"date-time":"2022-07-31T23:37:29Z","timestamp":1659310649000},"page":"1057","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Multivariate Time Series Imputation: An Approach Based on Dictionary Learning"],"prefix":"10.3390","volume":"24","author":[{"given":"Xiaomeng","family":"Zheng","sequence":"first","affiliation":[{"name":"Department of Statistics, University of Auckland, Auckland 1142, New Zealand"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4555-1714","authenticated-orcid":false,"given":"Bogdan","family":"Dumitrescu","sequence":"additional","affiliation":[{"name":"Department of Automatic Control and Computers, University Politehnica of Bucharest, 060042 Bucharest, Romania"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0824-0899","authenticated-orcid":false,"given":"Jiamou","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Computer Science, University of Auckland, Auckland 1142, New Zealand"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5512-0868","authenticated-orcid":false,"given":"Ciprian Doru","family":"Giurc\u0103neanu","sequence":"additional","affiliation":[{"name":"Department of Statistics, University of Auckland, Auckland 1142, New Zealand"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2022,7,31]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","unstructured":"Brockwell, P.J., and Davis, R.A. (2002). Introduction to Time Series and Forecasting, Springer.","DOI":"10.1007\/b97391"},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"768","DOI":"10.14778\/3377369.3377383","article-title":"Mind the gap: An experimental evaluation of imputation of missing values techniques in time series","volume":"13","author":"Khayati","year":"2020","journal-title":"VLDB Endow."},{"key":"ref_3","unstructured":"Li, L., Mccann, J., Pollard, N., and Faloutsos, C. (July, January 28). DynaMMo: Mining and summarization of coevolving sequences with missing values. Proceedings of the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Paris, France."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"520","DOI":"10.1093\/bioinformatics\/17.6.520","article-title":"Missing value estimation methods for DNA microarrays","volume":"17","author":"Troyanskaya","year":"2001","journal-title":"Bioinformatics"},{"key":"ref_5","first-page":"2287","article-title":"Spectral Regularization Algorithms for Learning Large Incomplete Matrices","volume":"11","author":"Mazumder","year":"2010","journal-title":"J. Mach. Learn. Res."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1956","DOI":"10.1137\/080738970","article-title":"A Singular Value Thresholding Algorithm for Matrix Completion","volume":"20","author":"Cai","year":"2010","journal-title":"SIAM J. Optim."},{"key":"ref_7","unstructured":"Khayati, M., B\u00f6hlen, M., and Gamper, J. (April, January 31). Memory-efficient centroid decomposition for long time series. Proceedings of the IEEE 30th International Conference on Data Engineering, Chicago, IL, USA."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"2257","DOI":"10.1007\/s10115-019-01421-7","article-title":"Scalable recovery of missing blocks in time series with high and low cross-correlations","volume":"62","author":"Khayati","year":"2020","journal-title":"Knowl. Inf. Syst."},{"key":"ref_9","first-page":"1460","article-title":"Global Convergence of a Grassmannian Gradient Descent Algorithm for Subspace Estimation","volume":"Volume 51","author":"Gretton","year":"2016","journal-title":"Proceedings of the 19th International Conference on Artificial Intelligence and Statistics"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1293","DOI":"10.1109\/JPROC.2018.2847041","article-title":"Streaming PCA and Subspace Tracking: The Missing Data Case","volume":"106","author":"Balzano","year":"2018","journal-title":"Proc. IEEE"},{"key":"ref_11","first-page":"11.1","article-title":"Robust principal component analysis?","volume":"58","author":"Li","year":"2011","journal-title":"J. ACM"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Shu, X., Porikli, F., and Ahuja, N. (2014, January 23\u201328). Robust Orthonormal Subspace Learning: Efficient Recovery of Corrupted Low-Rank Matrices. Proceedings of the Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA.","DOI":"10.1109\/CVPR.2014.495"},{"key":"ref_13","unstructured":"Mei, J., Castro, Y.D., Goude, Y., and H\u00e9brail, G. (2017, January 6\u201311). Nonnegative Matrix Factorization for Time Series Recovery From a Few Temporal Aggregates. Proceedings of the 34th International Conference on Machine Learning, Sydney, Australia."},{"key":"ref_14","unstructured":"Yu, H., Rao, N., and Dhillon, I.S. (2016, January 5\u201310). Temporal Regularized Matrix Factorization for High-dimensional Time Series Prediction. Proceedings of the Advances in Neural Information Processing Systems 29: Annual Conference on Neural Information Processing Systems, Barcelona, Spain."},{"key":"ref_15","doi-asserted-by":"crossref","unstructured":"Zheng, X., Dumitrescu, B., Liu, J., and Giurc\u0103neanu, C.D. (2021, January 23\u201327). On the Use of Dictionary Learning in Time Series Imputation. Proceedings of the 28th European Signal Processing Conference (EUSIPCO), Dublin, Ireland.","DOI":"10.23919\/Eusipco47968.2020.9287458"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Dumitrescu, B., and Irofti, P. (2018). Dictionary Learning Algorithms and Applications, Springer.","DOI":"10.1007\/978-3-319-78674-2"},{"key":"ref_17","unstructured":"Rubinstein, R., Zibulevsky, M., and Elad, M. (2008). Efficient Implementation of the K-SVD Algorithm Using Batch Orthogonal Matching Pursuit, Technion Univ."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"883","DOI":"10.1109\/TSP.2013.2295062","article-title":"Learning Overcomplete Dictionaries Based on Atom-by-Atom Updating","volume":"62","author":"Sadeghi","year":"2014","journal-title":"IEEE Trans. Signal Process."},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"366","DOI":"10.1109\/TCI.2018.2840334","article-title":"Convolutional dictionary learning: A comparative review and new algorithms","volume":"4","author":"Wohlberg","year":"2018","journal-title":"IEEE Trans. Comput. Imaging"},{"key":"ref_20","unstructured":"Guichaoua, C., and Dictionary Learning for Audio Inpainting (2020, August 04). Report HAL Id: Dumas-00725263, HAL Archives-Ouvertes. Available online: http:\/\/dumas.ccsd.cnrs.fr\/dumas-00725263."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1109\/TIP.2007.911828","article-title":"Sparse representation for color image restoration","volume":"17","author":"Mairal","year":"2008","journal-title":"IEEE Trans. Image Process."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"214","DOI":"10.1137\/070697653","article-title":"Learning multiscale sparse representations for image and video restoration","volume":"7","author":"Mairal","year":"2008","journal-title":"Multiscale Model. Simul."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"1839","DOI":"10.1109\/LSP.2016.2616354","article-title":"Large inpainting of face images with Trainlets","volume":"23","author":"Sulam","year":"2016","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"2862","DOI":"10.1016\/j.ijleo.2014.01.003","article-title":"Dictionary learning based sinogram inpainting for CT sparse reconstruction","volume":"125","author":"Li","year":"2014","journal-title":"Optik"},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"1025","DOI":"10.1109\/JSTSP.2011.2157892","article-title":"Fast dictionary learning for sparse representations of speech signals","volume":"5","author":"Jafari","year":"2011","journal-title":"IEEE J. Sel. Top. Signal Process."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Dumitrescu, B., and Giurc\u0103neanu, C.D. (2019). Adaptive-size dictionary learning using Information Theoretic Criteria. Algorithms, 12.","DOI":"10.3390\/a12090178"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"631","DOI":"10.1109\/LSP.2012.2209871","article-title":"Stagewise K-SVD to design efficient dictionaries for sparse representations","volume":"19","author":"Rusu","year":"2012","journal-title":"IEEE Signal Process. Lett."},{"key":"ref_28","doi-asserted-by":"crossref","unstructured":"Mazhar, R., and Gader, P.D. (2008, January 8\u201311). EK-SVD: Optimized Dictionary Design for Sparse Representations. Proceedings of the 19th International Conference on Pattern Recognition, Tampa, FL, USA.","DOI":"10.1109\/ICPR.2008.4761362"},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"130","DOI":"10.1109\/TIP.2011.2160072","article-title":"Nonparametric Bayesian dictionary learning for analysis of noisy and incomplete images","volume":"21","author":"Zhou","year":"2012","journal-title":"IEEE Trans. Image Process."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"221","DOI":"10.1007\/s11265-016-1154-1","article-title":"Towards Dictionaries of Optimal Size: A Bayesian Non Parametric Approach","volume":"90","author":"Dang","year":"2018","journal-title":"J. Signal Process. Syst."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"108488","DOI":"10.1016\/j.sigpro.2022.108488","article-title":"Dictionary learning for signals in additive noise with generalized Gaussian distribution","volume":"195","author":"Zheng","year":"2022","journal-title":"Signal Process."},{"key":"ref_32","unstructured":"Pati, Y., Rezaiifar, R., and Krishnaprasad, P. (1993, January 1\u20133). Orthogonal matching pursuit: Recursive function approximation with applications to wavelet decomposition. Proceedings of the 27th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, USA."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"60","DOI":"10.1109\/TSP.2015.2477047","article-title":"Outlier-Robust Greedy Pursuit Algorithms in \u2113p-Space for Sparse Approximation","volume":"64","author":"Zeng","year":"2016","journal-title":"IEEE Trans. Signal Process."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"461","DOI":"10.1214\/aos\/1176344136","article-title":"Estimating the Dimension of a Model","volume":"6","author":"Schwarz","year":"1978","journal-title":"Ann. Stat."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"759","DOI":"10.1093\/biomet\/asn034","article-title":"Extended Bayesian Information Criteria for Model Selection with Large Model Spaces","volume":"95","author":"Chen","year":"2008","journal-title":"Biometrika"},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"2035","DOI":"10.1109\/18.340476","article-title":"A Communication Channel Modeled on Contagion","volume":"40","author":"Alajaji","year":"1994","journal-title":"IEEE Trans. Inf. Theory"},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"230","DOI":"10.1109\/18.481793","article-title":"Detection of binary Markov sources over channels with additive Markov noise","volume":"42","author":"Alajaji","year":"1996","journal-title":"IEEE Trans. Inf. Theory"},{"key":"ref_38","doi-asserted-by":"crossref","first-page":"668","DOI":"10.1109\/26.592605","article-title":"Quantization of Memoryless and Gauss-Markov Sources Over Binary Markov Channels","volume":"45","author":"Phamdo","year":"1997","journal-title":"IEEE Trans. Commun."},{"key":"ref_39","doi-asserted-by":"crossref","first-page":"485","DOI":"10.1007\/s00034-002-0705-9","article-title":"Robust Index Assignment Using Hadamard Transform for Vector Quantization Transmission over Finite-Memory Contagion Channels","volume":"21","author":"Iordache","year":"2002","journal-title":"Circuits Syst. Signal Process."},{"key":"ref_40","doi-asserted-by":"crossref","first-page":"1243","DOI":"10.1109\/83.784436","article-title":"Image Segmentation and Labeling Using the Polya Urn Model","volume":"8","author":"Banerjee","year":"1999","journal-title":"IEEE Trans. Image Process."},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"1998","DOI":"10.1109\/TCNS.2017.2781467","article-title":"A Polya Contagion Model for Networks","volume":"5","author":"Hayhoe","year":"2018","journal-title":"IEEE Trans. Control Netw. Syst."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"207","DOI":"10.32614\/RJ-2017-009","article-title":"imputeTS: Time Series Missing Value Imputation in R","volume":"9","author":"Moritz","year":"2017","journal-title":"R J."},{"key":"ref_43","unstructured":"Moritz, S., and Gatwa, S. (2022, June 16). Package: ImputeTS, Version 3.0. Available online: https:\/\/cran.r-project.org\/web\/packages\/imputeTS\/index.html."},{"key":"ref_44","doi-asserted-by":"crossref","first-page":"193","DOI":"10.1007\/s100510050929","article-title":"Hierarchical structure in financial markets","volume":"11","author":"Mantegna","year":"1999","journal-title":"Eur. Phys. J. B"}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/24\/8\/1057\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,11]],"date-time":"2025-10-11T00:00:24Z","timestamp":1760140824000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/24\/8\/1057"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,7,31]]},"references-count":44,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2022,8]]}},"alternative-id":["e24081057"],"URL":"https:\/\/doi.org\/10.3390\/e24081057","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,7,31]]}}}