Improve Session-Based Recommendation with Triplet Mining and Dynamic Perturbations Graph Neural Networks

Jiayi ZHU, Yong FENG*, Mingliang ZHOU, Xiancai XIONG, Yongheng WANG, Yu XIA, Baohua QIANG, Qin MAO*, Bin FANG

*Corresponding author for this work

Research output: Journal PublicationsJournal Article (refereed)peer-review


Session-based recommendation (SBR) emphasizes mining user interests to predict the next click based on recent interactions within sessions. Most current SBR methods suffer from insufficient interactive information problems and fail to distinguish session representations with high similarities, which can neglect the inherent features within sessions. To fill the gap, we propose a triplet mining enhanced graph neural networks (TME-GNN) approach to enhance the recommendation systems by mining structural and inherent information. Technically, we first generate anchor, positive and negative embeddings based on the given session and set a triplet mining task to improve the recommendation task with subtle features by pushing positive pairs close and pulling negative pairs away. Second, to robust the model, we employ a self-supervised auxiliary task by adding dynamic perturbations to the embedding space. We conduct extensive experiments to demonstrate the superiority of our method against other state-of-the-art algorithms.

Original languageEnglish
Article number2350012
JournalInternational Journal of Pattern Recognition and Artificial Intelligence
Issue number6
Early online date14 Apr 2023
Publication statusPublished - May 2023
Externally publishedYes

Bibliographical note

Publisher Copyright:
© 2023 World Scientific Publishing Company.


  • collaborative filtering
  • graph learning
  • information retrieval
  • Recommendation system
  • session-based recommendation
  • triplet mining


Dive into the research topics of 'Improve Session-Based Recommendation with Triplet Mining and Dynamic Perturbations Graph Neural Networks'. Together they form a unique fingerprint.

Cite this