EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Transformer-Based Multi-Behavior Sequential Recommendation Method

Weining Bai, Bo He, Xia Yu, Yunfei Chen

Journal of Computer Science and Artificial Intelligence · 28.09.2026 · 8 (3) · 49-54

DOI: 10.54097/06t11748 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Computer Science and Artificial Intelligence, online erschienen am 28.09.2026. Verfasst von Weining Bai und 3 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Multi-Behavior Sequential Recommendation (MBSR) is an important research direction in sequential recommendation. It aims to exploit auxiliary behaviors in users’ historical interactions to improve item prediction under a target behavior. In recent years, Transformer architectures have provided an effective foundation for MBSR because self-attention can model long-range sequential dependencies and interactions among heterogeneous behaviors. This paper reviews recent advances in Transformer-based MBSR. We first introduce the task formulation, research background, and representative application scenarios. Representative methods are then reviewed in terms of behavior representation, relation modeling, temporal modeling, and interest modeling, with emphasis on how auxiliary behaviors are used to improve target-behavior item prediction and characterize evolving user interests. Finally, we summarize the main challenges, with particular attention to noisy auxiliary behaviors, efficient long-sequence modeling, and coordination among multiple target behaviors, and discuss promising directions for future research.

Zitieren

APA

Bai, W., He, B., Yu, X., & Chen, Y. (2026). Transformer-Based Multi-Behavior Sequential Recommendation Method. Journal of Computer Science and Artificial Intelligence, 8(3), 49-54. https://doi.org/10.54097/06t11748

Vancouver

Bai W, He B, Yu X, Chen Y. Transformer-Based Multi-Behavior Sequential Recommendation Method. Journal of Computer Science and Artificial Intelligence. 2026;8(3):49-54. doi:10.54097/06t11748

BibTeX
@article{Bai2026,
  author = {Bai, Weining and He, Bo and Yu, Xia and Chen, Yunfei},
  title = {Transformer-Based Multi-Behavior Sequential Recommendation Method},
  journal = {Journal of Computer Science and Artificial Intelligence},
  year = {2026},
  volume = {8},
  number = {3},
  pages = {49--54},
  doi = {10.54097/06t11748}
}
Kostenlos starten →