Recurrent knowledge graph embedding for effective recommendation

Zhu Sun, Jie Yang, Jie Zhang, Alessandro Bozzon, Long Kai Huang, Chi Xu

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

262 Citations (Scopus)
362 Downloads (Pure)


Knowledge graphs (KGs) have proven to be effective to improve recommendation. Existing methods mainly rely on hand-engineered features from KGs (e.g., meta paths), which requires domain knowledge. This paper presents RKGE, a KG embedding approach that automatically learns semantic representations of both entities and paths between entities for characterizing user preferences towards items. Specifically, RKGE employs a novel recurrent network architecture that contains a batch of recurrent networks to model the semantics of paths linking a same entity pair, which are seamlessly fused into recommendation. It further employs a pooling operator to discriminate the saliency of different paths in characterizing user preferences towards items. Extensive validation on real-world datasets shows the superiority of RKGE against state-of-the-art methods. Furthermore, we show that RKGE provides meaningful explanations for recommendation results.

Original languageEnglish
Title of host publicationRecSys '18
Subtitle of host publicationProceedings of the 12th ACM Conference on Recommender Systems
Place of PublicationNew York, NY
PublisherAssociation for Computer Machinery
Number of pages9
ISBN (Print)978-1-4503-5901-6
Publication statusPublished - 2018
Event12th ACM Conference on Recommender Systems, RecSys 2018 - Vancouver, Canada
Duration: 2 Oct 20187 Oct 2018


Conference12th ACM Conference on Recommender Systems, RecSys 2018

Bibliographical note

Accepted author manuscript


  • Attention Mechanism
  • Knowledge Graph
  • Recurrent Neural Network
  • Semantic Representation


Dive into the research topics of 'Recurrent knowledge graph embedding for effective recommendation'. Together they form a unique fingerprint.

Cite this