File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: SUMSHINE: Scalable Unsupervised Multi-Source Heterogeneous Information Network Embeddings

TitleSUMSHINE: Scalable Unsupervised Multi-Source Heterogeneous Information Network Embeddings
Authors
KeywordsAdversarial Learning
Distribution alignment
Graph representation learning
recommendation system
Issue Date30-Nov-2022
PublisherIEEE
Abstract

Heterogeneous information networks (HINs) have been extensively applied to real-world tasks, such as recommendation systems, social networks, and citation networks. While existing HIN representation learning methods can effectively learn the semantic and structural features in the network, little awareness was given to the distribution discrepancy of subgraphs within a single HIN. However, we find that ignoring such distribution discrepancy among subgraphs from multiple sources would hinder the effectiveness of graph embedding learning algorithms. This motivates us to propose SUMSHINE (Scalable Unsupervised Multi-Source Heterogeneous Information Network Embedding) — a scalable unsupervised framework to align the embedding distributions among multiple sources of an HIN. Experimental results on real-world datasets in a variety of downstream tasks validate the performance of our method over the state-of-the-art heterogeneous information network embedding algorithms.


Persistent Identifierhttp://hdl.handle.net/10722/337587
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorChan, TH-
dc.contributor.authorWong, CH-
dc.contributor.authorShen, J-
dc.contributor.authorYin, G-
dc.date.accessioned2024-03-11T10:22:17Z-
dc.date.available2024-03-11T10:22:17Z-
dc.date.issued2022-11-30-
dc.identifier.urihttp://hdl.handle.net/10722/337587-
dc.description.abstract<p>Heterogeneous information networks (HINs) have been extensively applied to real-world tasks, such as recommendation systems, social networks, and citation networks. While existing HIN representation learning methods can effectively learn the semantic and structural features in the network, little awareness was given to the distribution discrepancy of subgraphs within a single HIN. However, we find that ignoring such distribution discrepancy among subgraphs from multiple sources would hinder the effectiveness of graph embedding learning algorithms. This motivates us to propose SUMSHINE (Scalable Unsupervised Multi-Source Heterogeneous Information Network Embedding) — a scalable unsupervised framework to align the embedding distributions among multiple sources of an HIN. Experimental results on real-world datasets in a variety of downstream tasks validate the performance of our method over the state-of-the-art heterogeneous information network embedding algorithms.<br></p>-
dc.languageeng-
dc.publisherIEEE-
dc.relation.ispartofIEEE International Conference on Knowledge Graph (30/11/2022-01/12/2022, virtual)-
dc.subjectAdversarial Learning-
dc.subjectDistribution alignment-
dc.subjectGraph representation learning-
dc.subjectrecommendation system-
dc.titleSUMSHINE: Scalable Unsupervised Multi-Source Heterogeneous Information Network Embeddings-
dc.typeConference_Paper-
dc.identifier.doi10.1109/ickg55886.2022.00012-
dc.identifier.scopuseid_2-s2.0-85148540825-
dc.identifier.spage32-
dc.identifier.epage39-
dc.identifier.isiWOS:001050887100005-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats