A novel LSTM for multivariate time series with massive missingness

dc.contributorAalto-yliopistofi
dc.contributorAalto Universityen
dc.contributor.authorFouladgar, Nazaninen_US
dc.contributor.authorFrämling, Karyen_US
dc.contributor.departmentUmeå Universityen_US
dc.contributor.departmentFrämling Kary groupen_US
dc.contributor.departmentDepartment of Computer Scienceen
dc.date.accessioned2020-06-25T08:43:50Z
dc.date.available2020-06-25T08:43:50Z
dc.date.issued2020-05-02en_US
dc.description.abstractMultivariate time series with missing data is ubiquitous when the streaming data is collected by sensors or any other recording instruments. For instance, the outdoor sensors gathering different meteorological variables may encounter low material sensitivity to specific situations, leading to incomplete information gathering. This is problematic in time series prediction with massive missingness and different missing rate of variables. Contribution addressing this problem on the regression task of meteorological datasets by employing Long Short-Term Memory (LSTM), capable of controlling the information flow with its memory unit, is still missing. In this paper, we propose a novel model called forward and backward variable-sensitive LSTM (FBVS-LSTM) consisting of two decay mechanisms and some informative data. The model inputs are mainly the missing indicator, time intervals of missingness in both forward and backward direction and missing rate of each variable. We employ this information to address the so-called missing not at random (MNAR) mechanism. Separately learning the features of each parameter, the model becomes adapted to deal with massive missingness. We conduct our experiment on three real-world datasets for the air pollution forecasting. The results demonstrate that our model performed well along with other LSTM-derivation models in terms of prediction accuracy.en
dc.description.versionPeer revieweden
dc.format.extent17
dc.format.extent1-17
dc.format.mimetypeapplication/pdfen_US
dc.identifier.citationFouladgar , N & Främling , K 2020 , ' A novel LSTM for multivariate time series with massive missingness ' , Sensors (Switzerland) , vol. 20 , no. 10 , 2832 , pp. 1-17 . https://doi.org/10.3390/s20102832en
dc.identifier.doi10.3390/s20102832en_US
dc.identifier.issn1424-8220
dc.identifier.otherPURE UUID: ed44112c-31dd-4910-8acf-a50289e2f470en_US
dc.identifier.otherPURE ITEMURL: https://research.aalto.fi/en/publications/ed44112c-31dd-4910-8acf-a50289e2f470en_US
dc.identifier.otherPURE LINK: http://www.scopus.com/inward/record.url?scp=85084961927&partnerID=8YFLogxKen_US
dc.identifier.otherPURE FILEURL: https://research.aalto.fi/files/43455106/Fouladgar_A_Novel_LSTM.Sensors_20_02832_v2.pdfen_US
dc.identifier.urihttps://aaltodoc.aalto.fi/handle/123456789/45239
dc.identifier.urnURN:NBN:fi:aalto-202006254196
dc.language.isoenen
dc.publisherMultidisciplinary Digital Publishing Institute (MDPI)
dc.relation.ispartofseriesSensors (Switzerland)en
dc.relation.ispartofseriesVolume 20, issue 10en
dc.rightsopenAccessen
dc.subject.keywordLSTMen_US
dc.subject.keywordMassive missingnessen_US
dc.subject.keywordMultivariate time seriesen_US
dc.subject.keywordRegressionen_US
dc.titleA novel LSTM for multivariate time series with massive missingnessen
dc.typeA1 Alkuperäisartikkeli tieteellisessä aikakauslehdessäfi
dc.type.versionpublishedVersion
Files