{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,2]],"date-time":"2026-06-02T08:33:44Z","timestamp":1780389224782,"version":"3.54.1"},"reference-count":34,"publisher":"IOP Publishing","issue":"4","license":[{"start":{"date-parts":[[2020,10,15]],"date-time":"2020-10-15T00:00:00Z","timestamp":1602720000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2020,10,15]],"date-time":"2020-10-15T00:00:00Z","timestamp":1602720000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"name":"Department of Energy National Nuclear Security Administration","award":["DEAC52-06NA25396"],"award-info":[{"award-number":["DEAC52-06NA25396"]}]},{"name":"LANL Laboratory Directed Research and Development","award":["20190020DR"],"award-info":[{"award-number":["20190020DR"]}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2020,12,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Currently, high-dimensional data is ubiquitous in data science, which necessitates the development of techniques to decompose and interpret such multidimensional (aka tensor) datasets. Finding a low dimensional representation of the data, that is, its inherent structure, is one of the approaches that can serve to understand the dynamics of low dimensional latent features hidden in the data. Moreover, decomposition methods with non-negative constraints are shown to extract more insightful factors. Nonnegative RESCAL is one such technique, particularly well suited to analyze self-relational data, such as dynamic networks found in international trade flows. Particularly, non-negative RESCAL computes a low dimensional tensor representation by finding the latent space containing multiple modalities. Furthermore, estimating the dimensionality of this latent space is crucial for extracting meaningful latent features. Here, to determine the dimensionality of the latent space with non-negative RESCAL, we propose a latent dimension determination method which is based on clustering of the solutions of multiple realizations of non-negative RESCAL decompositions. We demonstrate the performance of our model selection method on synthetic data. We then apply our method to decompose a network of international trade flows data from <jats:italic>International Monetary Fund<\/jats:italic> and shows that with a correct latent dimension determination, the resulting features are able to capture relevant empirical facts from economic literature.<\/jats:p>","DOI":"10.1088\/2632-2153\/aba9ee","type":"journal-article","created":{"date-parts":[[2020,7,28]],"date-time":"2020-07-28T22:18:48Z","timestamp":1595974728000},"page":"045017","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":5,"title":["Determination of latent dimensionality in international trade flow"],"prefix":"10.1088","volume":"1","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9806-3274","authenticated-orcid":false,"given":"Duc P","family":"Truong","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7707-0838","authenticated-orcid":false,"given":"Erik","family":"Skau","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Vladimir I","family":"Valtchinov","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8636-4603","authenticated-orcid":false,"given":"Boian S","family":"Alexandrov","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2020,10,15]]},"reference":[{"key":"mlstaba9eebib1","doi-asserted-by":"publisher","first-page":"9","DOI":"10.1145\/1348549.1348551","article-title":"Exploiting time-varying relationships in statistical relational models","author":"Sharan","year":"2007"},{"key":"mlstaba9eebib2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.2200\/S00692ED1V01Y201601AIM032","volume":"10","author":"Raedt","year":"2016","journal-title":"Synthesis Lectures on Artificial Intelligence and Machine Learning"},{"key":"mlstaba9eebib3","doi-asserted-by":"publisher","first-page":"1195","DOI":"10.1145\/1989323.1989451","article-title":"Learning Statistical Models From Relational Data","author":"Getoor","year":"2011"},{"key":"mlstaba9eebib4","first-page":"175","article-title":"Latent Class Models","author":"Magidson","year":"2004"},{"key":"mlstaba9eebib5","first-page":"809","article-title":"A three-way model for collective learning on multi-relational data","volume":"11","author":"Nickel","year":"2011","journal-title":"Icml"},{"key":"mlstaba9eebib6","doi-asserted-by":"publisher","first-page":"pp 49","DOI":"10.1007\/978-3-319-22482-4_6","article-title":"Tensors and latent variable models","author":"Ishteva","year":"2015"},{"key":"mlstaba9eebib7","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1093\/imaiai\/iay009","volume":"8","author":"Robeva","year":"2019","journal-title":"Information and Inference: A Journal of the IMA"},{"key":"mlstaba9eebib8","author":"Koller","year":"2007"},{"key":"mlstaba9eebib9","author":"Everett","year":"2013"},{"key":"mlstaba9eebib10","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2019.2938890","article-title":"Hierarchical quality-relevant feature representation for soft sensor modeling: a novel deep learning strategy","author":"Yuan","year":"2019","journal-title":"IEEE Trans. Industrial Informatics"},{"key":"mlstaba9eebib11","doi-asserted-by":"publisher","first-page":"457","DOI":"10.1016\/j.isatra.2019.07.001","article-title":"A novel deep learning based fault diagnosis approach for chemical process with extended deep belief network","volume":"96","author":"Wang","year":"2020","journal-title":"ISA Trans."},{"key":"mlstaba9eebib12","author":"Cichocki","year":"2009"},{"key":"mlstaba9eebib13","doi-asserted-by":"publisher","first-page":"455","DOI":"10.1137\/07070111X","article-title":"Tensor decompositions and applications","volume":"51","author":"Kolda","year":"2009","journal-title":"SIAM Rev."},{"key":"mlstaba9eebib14","doi-asserted-by":"publisher","first-page":"788","DOI":"10.1038\/44565","article-title":"Learning the parts of objects by non-negative matrix factorization","volume":"401","author":"Lee","year":"1999","journal-title":"Nature"},{"key":"mlstaba9eebib15","first-page":"1053","article-title":"Bayesian nonlinear modeling for the prediction competition","volume":"100","author":"MacKay","year":"1994","journal-title":"ASHRAE Trans."},{"key":"mlstaba9eebib16","first-page":"2773","article-title":"Tensor decompositions for learning latent variable models","volume":"15","author":"Anandkumar","year":"2014","journal-title":"J. Machine Learning Res."},{"key":"mlstaba9eebib17","doi-asserted-by":"publisher","first-page":"644","DOI":"10.1016\/0196-6774(90)90014-6","article-title":"Tensor rank is NP-complete","volume":"11","author":"H\u00e5stad","year":"1990","journal-title":"J. Algorithms (Print)"},{"key":"mlstaba9eebib18","doi-asserted-by":"publisher","first-page":"352","DOI":"10.1002\/cem.1223","article-title":"Automatic relevance determination for multi-way models","volume":"23","author":"M\u00d8rup","year":"2009","journal-title":"J. Chemometrics: J. Chemometrics Soc."},{"key":"mlstaba9eebib19","first-page":"809","article-title":"Determining the number of latent factors in statistical multi-relational learning","volume":"20","author":"Shi","year":"2019","journal-title":"J. Machine Learning Res."},{"key":"mlstaba9eebib20","doi-asserted-by":"publisher","first-page":"3401","DOI":"10.1257\/aer.20101557","article-title":"Trade and the global recession","volume":"106","author":"Eaton","year":"2016","journal-title":"Am. Economic Rev."},{"key":"mlstaba9eebib21","doi-asserted-by":"publisher","first-page":"375","DOI":"10.1146\/annurev-economics-082912-110201","article-title":"The great trade collapse","volume":"5","author":"Bems","year":"2013","journal-title":"Annu. Rev. Econ."},{"key":"mlstaba9eebib22","article-title":"Nonnegative Canonical Polyadic Decomposition with Rank Deficient Factors","author":"Alexandrov","year":"2019"},{"key":"mlstaba9eebib23","doi-asserted-by":"publisher","first-page":"246","DOI":"10.1016\/j.celrep.2012.12.008","article-title":"Deciphering signatures of mutational processes operative in human cancer","volume":"3","author":"Alexandrov","year":"2013","journal-title":"Cell Rep."},{"key":"mlstaba9eebib24","doi-asserted-by":"publisher","first-page":"7332","DOI":"10.1002\/2013WR015037","article-title":"Blind source separation for groundwater pressure analysis based on nonnegative matrix factorization","volume":"50","author":"Alexandrov","year":"2014","journal-title":"Water Resour. Res."},{"key":"mlstaba9eebib25","doi-asserted-by":"publisher","first-page":"pp 1","DOI":"10.1007\/s11227-020-03181-6","article-title":"Distributed non-negative matrix factorization with determination of the number of latent features","author":"Chennupati","year":"2020","journal-title":"J. Supercomputing"},{"key":"mlstaba9eebib26","doi-asserted-by":"publisher","first-page":"415","DOI":"10.1038\/nature12477","article-title":"Signatures of mutational processes in human cancer","volume":"500","author":"Alexandrov","year":"2013","journal-title":"Nature"},{"key":"mlstaba9eebib27","doi-asserted-by":"publisher","first-page":"53","DOI":"10.1016\/0377-0427(87)90125-7","article-title":"Silhouettes: a graphical aid to the interpretation and validation of cluster analysis","volume":"20","author":"Rousseeuw","year":"1987","journal-title":"J. Comput. Appl. Math."},{"key":"mlstaba9eebib28","doi-asserted-by":"publisher","first-page":"205","DOI":"10.1287\/mksc.1.2.205","article-title":"A model for the analysis of asymmetric data in marketing research","volume":"1","author":"Harshman","year":"1982","journal-title":"Market. Sci."},{"key":"mlstaba9eebib29","doi-asserted-by":"crossref","DOI":"10.2172\/900402","article-title":"Pattern analysis of directed graphs using DEDICOM: an application to Enron email","author":"Bader","year":"2006"},{"key":"mlstaba9eebib30","article-title":"Non-negative tensor factorization with rescal","author":"Krompa\u00df","year":"2013"},{"key":"mlstaba9eebib31","article-title":"New estimates for direction of trade statistics New Estimates for Direction of Trade Statistics Working Paper No. 18\/16","author":"Marini","year":"2018"},{"key":"mlstaba9eebib32","article-title":"Factor Models for High-Dimensional Dynamic Networks: with Application to International Trade Flow Time Series 1981-2015","author":"Chen","year":"2017"},{"key":"mlstaba9eebib33","doi-asserted-by":"publisher","first-page":"594","DOI":"10.1080\/07474938.2010.481556","article-title":"An Empirical Comparison of Machine Learning Models for Time Series Forecasting","volume":"29","author":"Nesreen","year":"2010","journal-title":"Econometric Rev."},{"key":"mlstaba9eebib34","first-page":"1","article-title":"Statistical and Machine Learning forecasting methods: Concerns and ways forward","volume":"13","author":"Spyros Makridakis Evangelos Spiliotis","year":"2018","journal-title":"PLoS One"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee","content-type":"text\/html","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,1,31]],"date-time":"2022-01-31T01:33:15Z","timestamp":1643592795000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/aba9ee"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,10,15]]},"references-count":34,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2020,10,15]]},"published-print":{"date-parts":[[2020,12,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/aba9ee","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,10,15]]},"assertion":[{"value":"Determination of latent dimensionality in international trade flow","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2020 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2020-02-28","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2020-07-28","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2020-10-15","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}