BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer

Explore this paper's citation graph

Summary

A sequential recommendation model called BERT4Rec is proposed, which employs the deep bidirectional self-attention to model user behavior sequences, and outperforms various state-of-the-art sequential models consistently.

Type
preprint
Published
2019-04-14
Cited by
3,468
References
61
Access
Open access

Keywords

Computer science, Transformer, Encoder, Sequence (biology), Benchmark (surveying)

References

Cited by

Related papers