File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: E-LSTM-D: A Deep Learning Framework for Dynamic Network Link Prediction

TitleE-LSTM-D: A Deep Learning Framework for Dynamic Network Link Prediction
Authors
KeywordsDynamic network
encoder-decoder
link prediction
long short-term memory (LSTM)
network embedding
Issue Date2021
Citation
IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2021, v. 51, n. 6, p. 3699-3712 How to Cite?
AbstractPredicting the potential relations between nodes in networks, known as link prediction, has long been a challenge in network science. However, most studies just focused on link prediction of static network, while real-world networks always evolve over time with the occurrence and vanishing of nodes and links. Dynamic network link prediction (DNLP) thus has been attracting more and more attention since it can better capture the evolution nature of networks, but still most algorithms fail to achieve satisfied prediction accuracy. Motivated by the excellent performance of long short-term memory (LSTM) in processing time series, in this article, we propose a novel encoder-LSTM-decoder (E-LSTM-D) deep learning model to predict dynamic links end to end. It could handle long-term prediction problems, and suits the networks of different scales with fine-tuned structure. To the best of our knowledge, it is the first time that LSTM, together with an encoder-decoder architecture, is applied to link prediction in dynamic networks. This new model is able to automatically learn structural and temporal features in a unified framework, which can predict the links that never appear in the network before. The extensive experiments show that our E-LSTM-D model significantly outperforms newly proposed DNLP methods and obtain the state-of-the-art results.
Persistent Identifierhttp://hdl.handle.net/10722/330704
ISSN
2023 Impact Factor: 8.6
2023 SCImago Journal Rankings: 3.992
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorChen, Jinyin-
dc.contributor.authorZhang, Jian-
dc.contributor.authorXu, Xuanheng-
dc.contributor.authorFu, Chenbo-
dc.contributor.authorZhang, Dan-
dc.contributor.authorZhang, Qingpeng-
dc.contributor.authorXuan, Qi-
dc.date.accessioned2023-09-05T12:13:25Z-
dc.date.available2023-09-05T12:13:25Z-
dc.date.issued2021-
dc.identifier.citationIEEE Transactions on Systems, Man, and Cybernetics: Systems, 2021, v. 51, n. 6, p. 3699-3712-
dc.identifier.issn2168-2216-
dc.identifier.urihttp://hdl.handle.net/10722/330704-
dc.description.abstractPredicting the potential relations between nodes in networks, known as link prediction, has long been a challenge in network science. However, most studies just focused on link prediction of static network, while real-world networks always evolve over time with the occurrence and vanishing of nodes and links. Dynamic network link prediction (DNLP) thus has been attracting more and more attention since it can better capture the evolution nature of networks, but still most algorithms fail to achieve satisfied prediction accuracy. Motivated by the excellent performance of long short-term memory (LSTM) in processing time series, in this article, we propose a novel encoder-LSTM-decoder (E-LSTM-D) deep learning model to predict dynamic links end to end. It could handle long-term prediction problems, and suits the networks of different scales with fine-tuned structure. To the best of our knowledge, it is the first time that LSTM, together with an encoder-decoder architecture, is applied to link prediction in dynamic networks. This new model is able to automatically learn structural and temporal features in a unified framework, which can predict the links that never appear in the network before. The extensive experiments show that our E-LSTM-D model significantly outperforms newly proposed DNLP methods and obtain the state-of-the-art results.-
dc.languageeng-
dc.relation.ispartofIEEE Transactions on Systems, Man, and Cybernetics: Systems-
dc.subjectDynamic network-
dc.subjectencoder-decoder-
dc.subjectlink prediction-
dc.subjectlong short-term memory (LSTM)-
dc.subjectnetwork embedding-
dc.titleE-LSTM-D: A Deep Learning Framework for Dynamic Network Link Prediction-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/TSMC.2019.2932913-
dc.identifier.scopuseid_2-s2.0-85106509422-
dc.identifier.volume51-
dc.identifier.issue6-
dc.identifier.spage3699-
dc.identifier.epage3712-
dc.identifier.eissn2168-2232-
dc.identifier.isiWOS:000652103000034-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats