Skip to main content
SpringerLink
Log in

Sequential recommendation with metric models based on frequent sequences

  • Published:
Data Mining and Knowledge Discovery Aims and scope Submit manuscript

Abstract

Modeling user preferences (long-term history) and user dynamics (short-term history) is of greatest importance to build efficient sequential recommender systems. The challenge lies in the successful combination of the whole user’s history and his recent actions (sequential dynamics) to provide personalized recommendations. Existing methods capture the sequential dynamics of a user using fixed-order Markov chains (usually first order chains) regardless of the user, which limits both the impact of the past of the user on the recommendation and the ability to adapt its length to the user profile. In this article, we propose to use frequent sequences to identify the most relevant part of the user history for the recommendation. The most salient items are then used in a unified metric model that embeds items based on user preferences and sequential dynamics. Extensive experiments demonstrate that our method outperforms state-of-the-art, especially on sparse datasets. We show that considering sequences of varying lengths improves the recommendations and we also emphasize that these sequences provide explanations on the recommendation.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6

Similar content being viewed by others

Notes

  1. https://bit.ly/3gwZAOF.

  2. http://grouplens.org/datasets/movielens/1m/.

  3. https://www.visiativ.com/en-us/.

  4. Only for Foursquare dataset, we observed that it is better not to have \(\gamma \) (remove \(\gamma \) and \((1-\gamma )\) in Equation 3). It is noteworthy that recommendations may be different to the case where \(\gamma =0.5\) due to the bias terms \(\beta _i\).

  5. Best hyperparameters for each dataset are reported in supplementary material (Lonjarret et al. 2020a).

  6. We only show HIT_25, HIT_50, NDCG_25 and NDCG_50 in the tables. HIT_5, HIT_10, NDCG_5 and NDCG_10 are reported in supplementary material (Lonjarret et al. 2020a).

  7. Performances of other nearest-neighbor-based approaches as Item-based KNN and Sequence-Aware Extensions (V-S-KNN, S-S-KNN and SF-S-KNN) are reported in supplementary material (Lonjarret et al. 2020a).

  8. To avoid running again the grid search, we took the best combination of hyperparameters that we previously found for k = 10.

  9. We only show HIT_25, HIT_50, NDCG_25 and NDCG_50 in the table. HIT_5, HIT_10, NDCG_5 and NDCG_10 are reported in supplementary material (Lonjarret et al. 2020a).

  10. The two numbers after the name of the dataset are respectively the value of \({\texttt {minCount}}\) and L.

  11. Column (A) and (B) are both equivalent to a first order Markov chain but we split it into two different columns to point out the fact that it is very uncommon that no item from F matches \(s_{u}^{[1,t]}\)

  12. Sequences that do not match any substring of F are omitted for columns (F) and (G).

References

  • Adomavicius G, Kwon Y (2012) Improving aggregate recommendation diversity using ranking-based techniques. IEEE Trans Knowl Data Eng 24(05):896–911. https://doi.org/10.1109/TKDE.2011.15

    Article  Google Scholar 

  • Aggarwal CC (2016) Recommender systems: the textbook, 1st edn. Springer, Berlin

    Book  Google Scholar 

  • Bayer I, He X, Kanagal B, Rendle S (2017) A generic coordinate descent framework for learning from implicit feedback. In: Proceedings of the 26th international conference on world wide web, international world wide web conferences steering committee, republic and canton of Geneva, CHE, WWW ’17, p 1341–1350, 10.1145/3038912.3052694

  • Bishop CM (2006) Pattern recognition and machine learning (information science and statistics). Springer, Berlin

    MATH  Google Scholar 

  • Chen S, Moore JL, Turnbull D, Joachims T (2012) Playlist prediction via metric embedding. In: Yang Q, Agarwal D, Pei J (eds) The 18th ACM SIGKDD international conference on knowledge discovery and data mining, KDD’12, Beijing, China, Aug 12-16, 2012, ACM, pp 714–722, 10.1145/2339530.2339643

  • Chen X, Xu H, Zhang Y, Tang J, Cao Y, Qin Z, Zha H (2018) Sequential recommendation with user memory networks. In: Proceedings of the eleventh ACM international conference on web search and data mining, ACM, New York, USA, WSDM ’18, pp 108–116, 10.1145/3159652.3159668

  • Devooght R, Bersini H (2017) Long and short-term recommendations with recurrent neural networks. In: Proceedings of the 25th conference on user modeling, adaptation and personalization, ACM, New York, USA, UMAP ’17, pp 13–21, 10.1145/3079628.3079670

  • Ding Y, Li X (2005) Time weight collaborative filtering. In: Proceedings of the 14th ACM international conference on information and knowledge management, ACM, New York, USA, CIKM ’05, pp 485–492, 10.1145/1099554.1099689

  • Feng S, Li X, Zeng Y, Cong G, Chee YM, Yuan Q (2015) Personalized ranking metric embedding for next new poi recommendation. In: Proceedings of the 24th international conference on artificial intelligence, AAAI Press, IJCAI’15, pp 2069–2075, http://dl.acm.org/citation.cfm?id=2832415.2832536

  • Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: Teh YW, Titterington M (eds) Proceedings of the thirteenth international conference on artificial intelligence and statistics, PMLR, Chia Laguna Resort, Sardinia, Italy, Proceedings of machine learning research, vol 9, pp 249–256, http://proceedings.mlr.press/v9/glorot10a.html

  • Gomez-Uribe CA, Hunt N (2016) The netflix recommender system: algorithms, business value, and innovation. ACM Trans Manag Inf Syst. https://doi.org/10.1145/2843948

    Article  Google Scholar 

  • Gulla JA, Zhang L, Liu P,Ozgobek O, Su X (2017) The adressa dataset for news recommendation. In: Proceedings of the international conference on web intelligence, association for computing machinery, New York, USA, WI ’17, p 1042–1048, 10.1145/3106426.3109436

  • Gusfield D (1997) Algorithms on strings trees and sequences computer science and computational biology. Cambridge University Press, UK

    Book  Google Scholar 

  • Harper FM, Konstan JA (2015) The movielens datasets: history and context. ACM Trans Interact Intell Syst. https://doi.org/10.1145/2827872

    Article  Google Scholar 

  • He R, McAuley JJ (2016) Fusing similarity models with markov chains for sparse sequential recommendation. In: Bonchi F, Domingo-Ferrer J, Baeza-Yates R, Zhou Z, Wu X (eds) IEEE 16th international conference on data mining, ICDM 2016, Dec 12-15, 2016, Barcelona, Spain, IEEE computer society, pp 191–200, 10.1109/ICDM.2016.0030

  • He R, Kang WC, McAuley J (2017a) Translation-based recommendation. In: Proceedings of the eleventh ACM conference on recommender systems, ACM, New York, USA, RecSys’17, pp 161–169, 10.1145/3109859.3109882

  • He X, Liao L, Zhang H, Nie L, Hu X, Chua T (2017b) Neural collaborative filtering. In: Barrett R, Cummings R, Agichtein E, Gabrilovich E (eds) Proceedings of the 26th International conference on world wide web, WWW 2017, Perth, Australia, Apr 3-7, 2017, ACM, pp 173–182, 10.1145/3038912.3052569

  • 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, Association for computing machinery, New York, USA, CIKM ’18, p 843–852, 10.1145/3269206.3271761

  • Hidasi B, Karatzoglou A, Baltrunas L, Tikk D (2016) Session-based recommendations with recurrent neural networks. In: Bengio Y, LeCun Y (eds) 4th international conference on learning representations, ICLR 2016, San Juan, Puerto Rico, May 2-4, 2016, Conference track proceedings, https://iclr.cc/archive/www/doku.php%3Fid=iclr2016:accepted-main.html

  • Huang J, Zhao WX, Dou H, Wen JR, Chang EY (2018) Improving sequential recommendation with knowledge-enhanced memory networks. In: The 41st International ACM SIGIR conference on research and development in information retrieval, ACM, New York, USA, SIGIR ’18, pp 505–514, 10.1145/3209978.3210017

  • Jannach D, Ludewig M (2017) When recurrent neural networks meet the neighborhood for session-based recommendation. In: Proceedings of the eleventh ACM conference on recommender systems, ACM, New York, USA, RecSys ’17, pp 306–310, 10.1145/3109859.3109872

  • Kabbur S, Ning X, Karypis G (2013) Fism: Factored item similarity models for top-n recommender systems. In: Proceedings of the 19th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, New York, USA, KDD’13, pp 659–667, 10.1145/2487575.2487589

  • Kang W, McAuley JJ (2018) Self-attentive sequential recommendation. In: IEEE international conference on data mining, ICDM 2018, Singapore, Nov 17-20, 2018, IEEE computer society, pp 197–206, 10.1109/ICDM.2018.00035

  • Kingma DP, Ba J (2015) Adam: A method for stochastic optimization. In: Bengio Y, LeCun Y (eds) 3rd international conference on learning representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference track proceedings

  • Koren Y (2009) Collaborative filtering with temporal dynamics. In: Proceedings of the 15th ACM SIGKDD international conference on knowledge discovery and data mining, ACM, New York, USA, KDD ’09, pp 447–456, 10.1145/1557019.1557072

  • Koren Y, Bell RM (2015) Advances in collaborative filtering. In: Recommender systems handbook, Springer, pp 77–118

  • Lonjarret C, Auburtin R, Robardet C, Plantevit M (2020a) REBUS: Supplementary materials, source code and datasets. https://bit.ly/3gwZAOF

  • Lonjarret C, Robardet C, Plantevit M, Auburtin R, Atzmueller M (2020b) Why should i trust this item? explaining the recommendations of any model. In: 2020 IEEE international conference on data science and advanced analytics (DSAA), IEEE, pp 526–535

  • Ludewig M, Jannach D (2018) Evaluation of session-based recommendation algorithms. User Model User-Adapt Interact 28(4–5):331–390. https://doi.org/10.1007/s11257-018-9209-6

    Article  Google Scholar 

  • McAuley J, Targett C, Shi Q, van den Hengel A (2015) Image-based recommendations on styles and substitutes. In: Proceedings of the 38th international ACM SIGIR conference on research and development in information retrieval, association for computing machinery, New York, USA, SIGIR ’15, p 43–52, 10.1145/2766462.2767755

  • Ning X, Karypis G (2011) Slim: Sparse linear methods for top-n recommender systems. In: Proceedings of the 2011 IEEE 11th international conference on data mining, IEEE computer society, Washington, DC, USA, ICDM’11, pp 497–506, 10.1109/ICDM.2011.134

  • Pasricha R, McAuley J (2018) Translation-based factorization machines for sequential recommendation. In: Proceedings of the 12th ACM conference on recommender systems, ACM, New York, USA, RecSys ’18, pp 63–71, 10.1145/3240323.3240356

  • Quadrana M, Cremonesi P, Jannach D (2018) Sequence-aware recommender systems. ACM Comput Surv. https://doi.org/10.1145/3190616

    Article  Google Scholar 

  • Rendle S (2012) Factorization machines with libfm. ACM Trans Intell Syst Technol 10(1145/2168752):2168771

    Google Scholar 

  • Rendle S, Freudenthaler C, Gantner Z, Schmidt-Thieme L (2009) Bpr: Bayesian personalized ranking from implicit feedback. In: Proceedings of the twenty-fifth conference on uncertainty in artificial intelligence, AUAI Press, Arlington, Virginia, United States, UAI ’09, pp 452–461, http://dl.acm.org/citation.cfm?id=1795114.1795167

  • 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, New York, USA, WWW’10, pp 811–820, 10.1145/1772690.1772773

  • Resnick P, Varian HR (1997) Recommender systems—introduction to the special section. Commun ACM 40(3):56–58. https://doi.org/10.1145/245108.245121

    Article  Google Scholar 

  • Said A, Bellogín A (2014) Comparative recommender system evaluation: benchmarking recommendation frameworks. In: Proceedings of the 8th ACM conference on recommender systems, association for computing machinery, New York, USA, RecSys ’14, p 129–136, 10.1145/2645710.2645746

  • Sanchez P, Bellogín A (2020) Time and sequence awareness in similarity metrics for recommendation. Inf Process Manag 57:102228. https://doi.org/10.1016/j.ipm.2020.102228

    Article  Google Scholar 

  • Sarwar B, Karypis G, Konstan J, Riedl J (2001) Item-based collaborative filtering recommendation algorithms. In: Proceedings of the 10th international conference on world wide web, ACM, New York, USA, WWW ’01, pp 285–295, 10.1145/371920.372071

  • Sedhain S, Menon AK, Sanner S, Xie L (2015) Autorec: Autoencoders meet collaborative filtering. In: Proceedings of the 24th international conference on world wide web, ACM, New York, USA, WWW ’15 Companion, pp 111–112, 10.1145/2740908.2742726

  • 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, New York, USA, WSDM ’18, pp 565–573, 10.1145/3159652.3159656

  • Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser L, Polosukhin I (2017) Attention is all you need. In: Guyon I, von Luxburg U, Bengio S, Wallach HM, Fergus R, Vishwanathan SVN, Garnett R (eds) Advances in Neural Information Processing Systems 30: Annual conference on neural information processing systems 2017, 4-9 Dec 2017, Long Beach, CA, USA, pp 5998–6008, http://papers.nips.cc/paper/7181-attention-is-all-you-need

  • Wang S, Cao L, Wang Y (2019) A survey on session-based recommender systems. CoRR arxiv: abs/1902.04864

  • Wu CY, Ahmed A, Beutel A, Smola AJ, Jing H (2017) Recurrent recommender networks. In: Proceedings of the tenth ACM international conference on web search and data mining, ACM, New York, NY, USA, WSDM ’17, pp 495–503, 10.1145/3018661.3018689

  • Xiong L, Chen X, Huang T, Schneider JG, Carbonell JG (2010) Temporal collaborative filtering with bayesian probabilistic tensor factorization. In: Proceedings of the SIAM international conference on data mining, SDM 2010, Apr 29 - May 1, 2010, Columbus, Ohio, USA, SIAM, pp 211–222, 10.1137/1.9781611972801.19

  • Yoshida T, Takeuchi I, Karasuyama M (2019) Learning interpretable metric between graphs: convex formulation and computation with graph mining. In: Proceedings of the 25th ACM SIGKDD international conference on knowledge discovery and data mining, association for computing machinery, New York, USA, KDD ’19, p 1026–1036, 10.1145/3292500.3330845

  • Zhang F, Yuan NJ, Lian D, Xie X, Ma WY (2016) Collaborative knowledge base embedding for recommender systems. In: Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, ACM, New York, NY, USA, KDD ’16, pp 353–362, 10.1145/2939672.2939673

  • Zhang S, Yao L, Sun A, Tay Y (2019) Deep learning based recommender system a survey and new perspectives. ACM Comput Surv. https://doi.org/10.1145/3285029

    Article  Google Scholar 

  • Zhao T, McAuley JJ, King I (2014) Leveraging social connections to improve personalized ranking for collaborative filtering. In: Li J, Wang XS, Garofalakis MN, Soboroff I, Suel T, Wang M (eds) Proceedings of the 23rd ACM international conference on conference on information and knowledge management, CIKM 2014, Shanghai, China, Nov 3-7, 2014, ACM, pp 261–270, 10.1145/2661829.2661998

Download references

Acknowledgements

This work was supported by the ACADEMICS grant of the IDEXLYON, project of the University of Lyon, PIA operated by ANR-16-IDEX-0005.

Author information

Authors and Affiliations

  1. Univ Lyon, INSA Lyon, CNRS, LIRIS UMR 5205, 69621, Villeurbanne, France

    Corentin Lonjarret & Céline Robardet

  2. Visiativ, 26 r Benoit Bennier, 69260, Charbonnières-les-Bains, France

    Corentin Lonjarret & Roch Auburtin

  3. Univ Lyon, Université de Lyon 1, CNRS, LIRIS UMR 5205, 69622, Villeurbanne, France

    Marc Plantevit

Authors
  1. Corentin Lonjarret
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Roch Auburtin
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Céline Robardet
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Marc Plantevit
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Céline Robardet.

Additional information

Responsible editor: Bart Goethals.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Lonjarret, C., Auburtin, R., Robardet, C. et al. Sequential recommendation with metric models based on frequent sequences. Data Min Knowl Disc 35, 1087–1133 (2021). https://doi.org/10.1007/s10618-021-00744-w

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10618-021-00744-w

Keywords

Search

Navigation