A Novel KNN Approach for Session-Based Recommendation

  • Huifeng Guo
  • Ruiming Tang
  • Yunming YeEmail author
  • Feng Liu
  • Yuzhou Zhang
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11440)


The KNN approach, which is widely used in recommender systems because of its efficiency, robustness and interpretability, is proposed for session-based recommendation recently and outperforms recurrent neural network algorithms. It captures the most recent co-occurrence information of items by considering the interaction time. However, it neglects the co-occurrence information of items in the historical behavior which is interacted earlier than others and cannot discriminate the impact of vertices with different popularity. Due to these observations, this paper presents a novel KNN approach to address these issues for session-based recommendation. Specifically, a diffusion-based similarity method is proposed for incorporating the popularity of items, and the candidate selection method is proposed to capture more co-occurrence information of items in the same session efficiently. Comprehensive experiments are conducted to demonstrate the effectiveness of our KNN approach over the state-of-the-art KNN approach for session-based recommendation on three benchmark datasets.


Diffusion model Session-based recommendation Nearest neighbor 



This research was supported in part by NSFC under Grant No. U1836107, and National Key R&D Program of China under Grant No. 2018YFB0504905.


  1. 1.
    Bonnin, G., Jannach, D.: Automated generation of music playlists: survey and experiments. ACM Comput. Surv. 47(2), 26:1–26:35 (2014)CrossRefGoogle Scholar
  2. 2.
    Covington, P., Jay, A., Sargin, E.: Deep neural networks for Youtube recommendations. In: ACM RecSys, pp. 191–198 (2016)Google Scholar
  3. 3.
    Guo, H., Tang, R., Ye, Y., Li, Z., He, X.: DeepFM: a factorization-machine based neural network for CTR prediction. In: IJCAI, pp. 1725–1731 (2017)Google Scholar
  4. 4.
    Guo, H., Tang, R., Ye, Y., Li, Z., He, X.: A graph-based push service platform. In: Candan, S., Chen, L., Pedersen, T.B., Chang, L., Hua, W. (eds.) DASFAA 2017. LNCS, vol. 10178, pp. 636–648. Springer, Cham (2017). Scholar
  5. 5.
    Hidasi, B., Karatzoglou, A., Baltrunas, L., Tikk, D.: Session-based recommendations with recurrent neural networks. CoRR abs/1511.06939 (2015)Google Scholar
  6. 6.
    Jannach, D., Ludewig, M.: When recurrent neural networks meet the neighborhood for session-based recommendation. In: ACM RecSys, pp. 306–310 (2017)Google Scholar
  7. 7.
    Kabbur, S., Ning, X., Karypis, G.: FISM: factored item similarity models for top-n recommender systems. In: SIGKDD, pp. 659–667 (2013)Google Scholar
  8. 8.
    Koren, Y., Bell, R.M., Volinsky, C.: Matrix factorization techniques for recommender systems. IEEE Comput. J. 42(8), 30–37 (2009)CrossRefGoogle Scholar
  9. 9.
    Ludewig, M., Jannach, D.: Evaluation of session-based recommendation algorithms. User Model. User-Adapt. Interact. 28(4–5), 331–390 (2018)CrossRefGoogle Scholar
  10. 10.
    Rendle, S., Freudenthaler, C., Schmidt-Thieme, L.: Factorizing personalized Markov chains for next-basket recommendation. In: WWW, pp. 811–820 (2010)Google Scholar
  11. 11.
    Yap, G., Li, X., Yu, P.S.: Effective next-items recommendation via personalized sequential pattern mining. In: DASFAA, pp. 48–64 (2012)Google Scholar
  12. 12.
    Zhou, T., Ren, J., Medo, M., Zhang, Y.: Bipartite network projection and personal recommendation. Phys. Rev. E 76(2), 046115 (2007)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Huifeng Guo
    • 2
  • Ruiming Tang
    • 2
  • Yunming Ye
    • 1
    • 3
    Email author
  • Feng Liu
    • 1
    • 3
  • Yuzhou Zhang
    • 2
  1. 1.Harbin Institute of TechnologyShenzhenChina
  2. 2.Noah’s Ark LabHuaweiChina
  3. 3.Shenzhen Key Laboratory of Internet Information CollaborationShenzhenChina

Personalised recommendations