Skip to main content
Erschienen in: Journal of Intelligent Information Systems 2/2024

06.11.2023 | Research

BMDF-SR: bidirectional multi-sequence decoupling fusion method for sequential recommendation

verfasst von: Aohua Gao, Jiwei Qin, Chao Ma, Tao Wang

Erschienen in: Journal of Intelligent Information Systems | Ausgabe 2/2024

Einloggen

Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.

search-config
loading …

Abstract

In the domain of sequence recommendation, contextual information has been shown to effectively improve the accuracy of predicting the user’s next interaction. However, existing studies do not consider the dependencies between contextual information and item sequences, but the contextual information is directly fusing with the item sequences, which brings the problems described below: (1) Direct fusion fuses contextual information (e.g., time and categories) with item sequences which increases the dimensionality of the embedding matrix, thus increasing the complexity of the attention computation. (2) The attention computation of heterogeneous context information in the same embedding matrix makes it difficult for the recommendation model to distinguish this heterogeneous information. Therefore, we propose a bidirectional multi-sequence decoupling fusion method for sequence recommendation (BMDF-SR) to address the above issues. To establish the dependencies between temporal context sequences and item sequences, we first treat temporal contextual information as independent sequences and build bidirectional dependencies between contextual information sequences and item sequences via a three-layer seq2seq structure. Then, we perform attention computation independently for context sequences such as categories, and the complexity of attention computation can be effectively reduced by this decoupled attention computation. Moreover, since the attention computation is performed separately for each sequence, the interference between heterogeneous information during sequence fusion is reduced, allowing the model to effectively discriminate between different types of information. Extensive experiments on four real-world datasets show that the BMDF-SR method outperforms popular models.

Sie haben noch keine Lizenz? Dann Informieren Sie sich jetzt über unsere Produkte:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Literatur
Zurück zum Zitat Hidasi, B., & Karatzoglou, A. (2018). Recurrent neural networks with top-k gains for session-based recommendations. In: Proceedings of the 27th ACM international conference on information and knowledge management. ACM, Torino, Italy. https://doi.org/10.1145/3269206.3271761 Hidasi, B., & Karatzoglou, A. (2018). Recurrent neural networks with top-k gains for session-based recommendations. In: Proceedings of the 27th ACM international conference on information and knowledge management. ACM, Torino, Italy. https://​doi.​org/​10.​1145/​3269206.​3271761
Zurück zum Zitat Hou, Y., Hu, B., & Zhang, Z., et al. (2022). Core: simple and effective session-based recommendation within consistent representation space. In: Proceedings of the 45th international ACM SIGIR conference on research and development in information retrieval. ACM, Madrid, Spain. https://doi.org/10.1145/3477495.3531955 Hou, Y., Hu, B., & Zhang, Z., et al. (2022). Core: simple and effective session-based recommendation within consistent representation space. In: Proceedings of the 45th international ACM SIGIR conference on research and development in information retrieval. ACM, Madrid, Spain. https://​doi.​org/​10.​1145/​3477495.​3531955
Zurück zum Zitat Le, D.T., Lauw, H.W., & Fang, Y. (2018) Modeling contemporaneous basket sequences with twin networks for next-item recommendation. In: Proceedings of the twenty-seventh international joint conference on artificial intelligence. IJCAI, Stockholm, Sweden. https://doi.org/10.24963/ijcai.2018/474 Le, D.T., Lauw, H.W., & Fang, Y. (2018) Modeling contemporaneous basket sequences with twin networks for next-item recommendation. In: Proceedings of the twenty-seventh international joint conference on artificial intelligence. IJCAI, Stockholm, Sweden. https://​doi.​org/​10.​24963/​ijcai.​2018/​474
Zurück zum Zitat Rendle, S., Freudenthaler, C., & Schmidt-Thieme, L. (2010). Factorizing personalized markov chains for next-basket recommendation. In: Proceedings of the 19th international conference on world wide web. ACM, Raleigh, North Carolina. https://doi.org/10.1145/1772690.1772773 Rendle, S., Freudenthaler, C., & Schmidt-Thieme, L. (2010). Factorizing personalized markov chains for next-basket recommendation. In: Proceedings of the 19th international conference on world wide web. ACM, Raleigh, North Carolina. https://​doi.​org/​10.​1145/​1772690.​1772773
Zurück zum Zitat Sun, F., Liu, J., & Wu, J., et al. (2019). Bert4rec: sequential recommendation with bidirectional encoder representations from transformer. In: Proceedings of the 28th ACM international conference on information and knowledge management. ACM, Beijing, China. https://doi.org/10.1145/3357384.3357895 Sun, F., Liu, J., & Wu, J., et al. (2019). Bert4rec: sequential recommendation with bidirectional encoder representations from transformer. In: Proceedings of the 28th ACM international conference on information and knowledge management. ACM, Beijing, China. https://​doi.​org/​10.​1145/​3357384.​3357895
Zurück zum Zitat Tang, J., & Wang, K. (2018). Personalized top-n sequential recommendation via convolutional sequence embedding. In: Proceedings of the eleventh ACM international conference on web search and data mining. ACM, Marina Del Rey, CA. https://doi.org/10.1145/3159652.3159656 Tang, J., & Wang, K. (2018). Personalized top-n sequential recommendation via convolutional sequence embedding. In: Proceedings of the eleventh ACM international conference on web search and data mining. ACM, Marina Del Rey, CA. https://​doi.​org/​10.​1145/​3159652.​3159656
Zurück zum Zitat Xie, Y., Zhou, P., & Kim, S. (2022). Decoupled side information fusion for sequential recommendation. In: Proceedings of the 45th international ACM SIGIR conference on research and development in information retrieval. ACM, Madrid, Spain. https://doi.org/10.1145/3477495.3531963 Xie, Y., Zhou, P., & Kim, S. (2022). Decoupled side information fusion for sequential recommendation. In: Proceedings of the 45th international ACM SIGIR conference on research and development in information retrieval. ACM, Madrid, Spain. https://​doi.​org/​10.​1145/​3477495.​3531963
Zurück zum Zitat Yuan, X., Duan, D., & Tong, L., et al. (2021). Icai-sr: Item categorical attribute integrated sequential recommendation. In: Proceedings of the 44th international ACM SIGIR conference on research and development in information retrieval. ACM, Virtual Event, Canada. https://doi.org/10.1145/3404835.3463060 Yuan, X., Duan, D., & Tong, L., et al. (2021). Icai-sr: Item categorical attribute integrated sequential recommendation. In: Proceedings of the 44th international ACM SIGIR conference on research and development in information retrieval. ACM, Virtual Event, Canada. https://​doi.​org/​10.​1145/​3404835.​3463060
Zurück zum Zitat Zhang, T., Zhao, P., & Liu, Y., et al. (2019). Feature-level deeper self-attention network for sequential recommendation. In: Proceedings of the twenty-eighth international joint conference on artificial intelligence. IJCAI, Macao, China. https://doi.org/10.24963/ijcai.2019/600 Zhang, T., Zhao, P., & Liu, Y., et al. (2019). Feature-level deeper self-attention network for sequential recommendation. In: Proceedings of the twenty-eighth international joint conference on artificial intelligence. IJCAI, Macao, China. https://​doi.​org/​10.​24963/​ijcai.​2019/​600
Zurück zum Zitat Zhao, W.X., Mu, S., & Hou, Y., et al. (2021). Recbole: Towards a unified, comprehensive and efficient framework for recommendation algorithms. In: Proceedings of the 30th ACM international conference on information & knowledge management. ACM, Virtual Event, Queensland. https://doi.org/10.1145/3459637.3482016 Zhao, W.X., Mu, S., & Hou, Y., et al. (2021). Recbole: Towards a unified, comprehensive and efficient framework for recommendation algorithms. In: Proceedings of the 30th ACM international conference on information & knowledge management. ACM, Virtual Event, Queensland. https://​doi.​org/​10.​1145/​3459637.​3482016
Zurück zum Zitat Zhou, K., Wang, H., & Zhao, W.X., et al. (2019). S3-rec: Self-supervised learning for sequential recommendation with mutual information maximization. In: Proceedings of the 29th ACM international conference on information & knowledge management. ACM, Virtual Event, Ireland. https://doi.org/10.1145/3340531.3411954 Zhou, K., Wang, H., & Zhao, W.X., et al. (2019). S3-rec: Self-supervised learning for sequential recommendation with mutual information maximization. In: Proceedings of the 29th ACM international conference on information & knowledge management. ACM, Virtual Event, Ireland. https://​doi.​org/​10.​1145/​3340531.​3411954
Metadaten
Titel
BMDF-SR: bidirectional multi-sequence decoupling fusion method for sequential recommendation
verfasst von
Aohua Gao
Jiwei Qin
Chao Ma
Tao Wang
Publikationsdatum
06.11.2023
Verlag
Springer US
Erschienen in
Journal of Intelligent Information Systems / Ausgabe 2/2024
Print ISSN: 0925-9902
Elektronische ISSN: 1573-7675
DOI
https://doi.org/10.1007/s10844-023-00825-w

Weitere Artikel der Ausgabe 2/2024

Journal of Intelligent Information Systems 2/2024 Zur Ausgabe

Premium Partner