Inproceedings,

BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer

, , , , , , and .
CIKM, page 1441-1450. ACM, (2019)
DOI: 10.1145/3357384.3357895

Abstract

Modeling users' dynamic preferences from their historical behaviors is challenging and crucial for recommendation systems. Previous methods employ sequential neural networks to encode users' historical interactions from left to right into hidden representations for making recommendations. Despite their effectiveness, we argue that such left-to-right unidirectional models are sub-optimal due to the limitations including: unidirectional architectures restrict the power of hidden representation in users' behavior sequences; they often assume a rigidly ordered sequence which is not always practical. To address these limitations, we proposed a sequential recommendation model called BERT4Rec, which employs the deep bidirectional self-attention to model user behavior sequences. To avoid the information leakage and efficiently train the bidirectional model, we adopt the Cloze objective to sequential recommendation, predicting the random masked items in the sequence by jointly conditioning on their left and right context. In this way, we learn a bidirectional representation model to make recommendations by allowing each item in user historical behaviors to fuse information from both left and right sides. Extensive experiments on four benchmark datasets show that our model outperforms various state-of-the-art sequential models consistently.

Tags

Users

  • @oriolc
  • @ruben_hussong
  • @jan_seidlein
  • @tobias.koopmann
  • @kherud
  • @s363405
  • @nosebrain
  • @dblp
  • @uw_ws20_ml

Comments and Reviewsshow / hide

  • @ruben_hussong
    @ruben_hussong 3 years ago
    Basismodell von Meantime, Vergleichsmodell von SASRec.
  • @kherud
    @kherud 3 years ago
    Bekannter Ansatz, der BERT auf sequenzielle Recommendation anwendet. Aufgenommen zum Vergleich mit eigener Methode basierend auf Graph Neural Networks.
  • @s363405
    @s363405 4 years ago
    Describes an extension to the SASRec model which utilizes the principles used in BERT.
Please log in to take part in the discussion (add own reviews or comments).