PPNW: personalized pairwise novelty loss weighting for novel recommendation

Abstract

Most works of recommender systems focus on providing users with highly accurate item predictions based on the assumption that accurate suggestions can best satisfy users. However, accuracy-focused models also create great system bias towards popular items and, as a result, unpopular items rarely get recommended and will stay as “cold items” forever. Both users and item providers will suffer in such scenario. To promote item novelty, which plays a crucial role in system robustness and diversity, previous studies focus mainly on re-ranking a top-N list generated by an accuracy-focused base model. The re-ranking algorithm is thus completely independent of the base model. Eventually, these frameworks are essentially limited by the base model and the separated 2 stages cause greater complication and inefficiency in providing novel suggestions. In this work, we propose a personalized pairwise novelty weighting framework for BPR loss function, which covers the limitations of BPR and effectively improves novelty with negligible decrease in accuracy. Base model will be guided by the novelty-aware loss weights to learn user preference and to generate novel top-N list in only 1 stage. Comprehensive experiments on 3 public datasets show that our approach effectively promotes novelty with almost no decrease in accuracy.

This is a preview of subscription content, access via your institution.

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

Notes

  1. 1.

    https://github.com/ArgentLo/PPNW-KAIS.

References

  1. 1.

    Adomavicius G, Kwon YO (2012) Improving aggregate recommendation diversity using ranking-based techniques. IEEE Trans Knowl Data Eng 24(5):896–911

    Article  Google Scholar 

  2. 2.

    Chen C-M et al (2019) Collaborative similarity embedding for recommender systems. In: The World Wide Web conference. ACM, pp 2637–2643

  3. 3.

    Cheng P et al (2017) Learning to recommend accurate and diverse items. In: Proceedings of the 26th international conference on World Wide Web. International World Wide Web Conferences Steering Committee, pp 183–192

  4. 4.

    Dong Q, Gong S, Zhu X (2019) Imbalanced deep learning by minority class incremental rectification. IEEE Trans Pattern Anal Mach Intell 41(6):1367–1381

    Article  Google Scholar 

  5. 5.

    Ebesu T, Shen B, Fang Y (2018) Collaborative memory network for recommendation systems. In: The 41st international ACM SIGIR conference on research & development in information retrieval. ACM, pp 515–524

  6. 6.

    Geng X et al (2015) Learning image and user features for recommendation in social networks. In: 2015 IEEE international conference on computer vision, ICCV 2015, Santiago, Chile, 7–13 Dec 2015, pp 4274–4282

  7. 7.

    Grover A, Leskovec J (2016) node2vec: scalable feature learning for networks. In: Proceedings of the 22nd ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, pp 855–864

  8. 8.

    Harald S (2010) Training and testing of recommender systems on data missing not at random. In: Proceedings of the 16th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, pp 713–722

  9. 9.

    Harper FM, Konstan JA (2016) The MovieLens datasets: history and context. TiiS 5(4):19:1–19:19. https://doi.org/10.1145/2827872

    Article  Google Scholar 

  10. 10.

    He X et al (2017) Neural collaborative filtering. In: Proceedings of the 26th international conference on World Wide Web. International World Wide Web Conferences Steering Committee, pp 173–182

  11. 11.

    He X et al (2016) Fast matrix factorization for online recommendation with implicit feedback. In: Proceedings of the 39th international ACM SIGIR conference on research and development in information retrieval, SIGIR 2016, Pisa, Italy, 17–21 July 2016, pp 549–558

  12. 12.

    Ho Y-C, Chiang Y-T, Hsu JY-J (2014) Who likes it more? Mining worth-recommending items from long tails by modeling relative preference. In: Proceedings of the 7th ACM international conference on web search and data mining. ACM, pp 253–262

  13. 13.

    Hu Y, Koren Y, Volinsky C (2008) Collaborative filtering for implicit feedback datasets. In: Proceedings of the 8th IEEE international conference on data mining (ICDM 2008), 15–19 Dec 2008, Pisa, Italy, pp 263–272

  14. 14.

    Hu Y, Koren Y, Volinsky C (2008) Collaborative filtering for implicit feedback datasets. In: 2008 Eighth IEEE international conference on data mining. IEEE, pp 263–272

  15. 15.

    Hurley N, Zhang M (2011) Novelty and diversity in top-N recommendation—analysis and evaluation. ACM Trans Internet Technol (TOIT) 10(4):14:1–14:30

    Article  Google Scholar 

  16. 16.

    Joachims T et al (2017) Accurately interpreting clickthrough data as implicit feedback. ACM SIGIR Forum, vol 51, no 1. ACM, New York, pp 4–11

  17. 17.

    Jugovac M, Jannach D, Lerche L (2017) Efficient optimization of multiple recommendation quality factors according to individual user tendencies. Expert Syst Appl 81:321–331

    Article  Google Scholar 

  18. 18.

    Kapoor K et al (2015) I like to explore sometimes: adapting to dynamic user novelty preferences. In: RecSys 2015, Vienna, Austria, 16–20 Sept 2015, pp 19–26

  19. 19.

    Kelly D, Teevan J (2003) Implicit feedback for inferring user preference: a bibliography. ACM SIGIR forum, vol 37, no 2. ACM, New York, pp 18–28

  20. 20.

    Kotkov D, Veijalainen J, Wang S (2016) Challenges of serendipity in recommender systems. In: Proceedings of the 12th international conference on web information systems and technologies, WEBIST 2016, vol 2, Rome, Italy, 23–25 April 2016, pp 251–256

  21. 21.

    Ma C et al (2018) Point-of-interest recommendation: exploiting selfattentive autoencoders with neighbor-aware influence. In: Proceedings of the 27th ACM international conference on information and knowledge management, CIKM 2018, Torino, Italy, 22–26 Oct, 2018. ACM, pp 697–706

  22. 22.

    Nakatsuji M et al (2010) Classical music for rock fans? Novel recommendations for expanding user interests. In: Proceedings of the 19th ACM international conference on information and knowledge management. ACM, pp 949–958

  23. 23.

    Ning Y et al (2017) A gradient-based adaptive learning framework for efficient personal recommendation. In: Proceedings of the eleventh ACM conference on recommender systems. ACM, pp 23–31

  24. 24.

    Oh J et al (2011) Novel recommendation based on personal popularity tendency. In: 2011 IEEE 11th international conference on data mining. IEEE, pp 507–516

  25. 25.

    Perozzi B, Al-Rfou R, Skiena S (2014) Deepwalk: online learning of social representations. In: Proceedings of the 20th ACM SIGKDD international conference on knowledge discovery and data mining. ACM, pp 701–710

  26. 26.

    Rendle S (2010) Factorization machines. In: ICDM 2010, The 10th IEEE international conference on data mining, Sydney, Australia, 14–17 Dec 2010, pp 995–1000

  27. 27.

    Rendle S et al (2009) BPR: Bayesian personalized ranking from implicit feedback. In: Proceedings of the twenty-fifth conference on uncertainty in artificial intelligence. AUAI Press, pp 452–461

  28. 28.

    Ricci F, Rokach L, Shapira B (2011) Introduction to recommender systems handbook. In: Recommender systems handbook, pp 1–35

  29. 29.

    Ricci F, Rokach L, Shapira B (2015) Recommender systems handbook. Springer, Berlin

    Google Scholar 

  30. 30.

    Steck H (2011) Item popularity and recommendation accuracy. In: Proceedings of the fifth ACM conference on recommender systems. ACM, pp 125–132

  31. 31.

    Tang J et al (2015) Line: large-scale information network embedding. In: Proceedings of the 24th international conference on world wide web. International World Wide Web Conferences Steering Committee, pp 1067–1077

  32. 32.

    Vargas S, Castells P (2014) Improving sales diversity by recommending users to items. In: Proceedings of the 8th ACM conference on recommender systems. ACM, pp 145–152

  33. 33.

    Vaswani A et al (2017) Attention is all you need. In: Advances in neural information processing systems, pp 5998–6008

  34. 34.

    Vijaikumar M, Shevade S, Murty MN (2019) SoRecGAT: leveraging graph attention mechanism for top-N social recommendation. In: Joint European conference on machine learning and knowledge discovery in databases. Springer, pp 430–446

  35. 35.

    Wang C, Blei DM (2011) Collaborative topic modeling for recommending scientific articles. In: Proceedings of the 17th ACM SIGKDD international conference on knowledge discovery and data mining, San Diego, CA, USA, 21–24 Aug 2011, pp 448–456

  36. 36.

    Wang Q et al (2018) Neural memory streaming recommender networks with adversarial training. In: Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining, KDD 2018, London, UK, 19–23 Aug 2018, pp 2467–2475

  37. 37.

    Wang X et al (2019) KGAT: knowledge graph attention network for recommendation. arXiv preprint arXiv:1905.07854

  38. 38.

    Wang X et al (2019) Neural graph collaborative filtering. arXiv preprint arXiv:1905.08108

  39. 39.

    Yang J-H et al (2018) HOP-rec: high-order proximity for implicit recommendation. In: Proceedings of the 12th ACM conference on recommender systems. ACM, pp 140–144

  40. 40.

    Yin H et al (2012) Challenging the long tail recommendation. PVLDB 5(9):896–907

    Google Scholar 

  41. 41.

    Yue S (2017) Imbalanced malware images classification: a CNN based approach. In: CoRR abs/1708.08042

  42. 42.

    Zhang F et al (2015) A novelty-seeking based dining recommender system. In: Proceedings of the 24th international conference on World Wide Web, WWW 2015, Florence, Italy, 18–22 May 2015, pp 1362–1372

  43. 43.

    Zhang M, Hurley N (2008) Avoiding monotony: improving the diversity of recommendation lists. In: Proceedings of the 2008 ACM conference on recommender systems, RecSys 2008, Lausanne, Switzerland, 23–25 Aug 2008, pp 123–130

  44. 44.

    Zhang YC et al (2012) Auralist: introducing serendipity into music recommendation. In: Proceedings of the fifth international conference on web search and web data mining, WSDM 2012, Seattle, WA, USA, 8–12 Feb 2012, pp 13–22

  45. 45.

    Zhang YC et al (2012) Auralist: introducing serendipity into music recommendation. In: Proceedings of the fifth international conference on web search and web data mining, WSDM 2012, Seattle, WA, USA, February 8-12, 2012, pp 13–22

  46. 46.

    Zhao P, Lee DL (2016) How much novelty is relevant? It depends on your curiosity. In: Proceedings of the 39th international ACM SIGIR conference on research and development in information retrieval, SIGIR 2016, Pisa, Italy, 17–21 July 2016, pp 315–324

  47. 47.

    Ziegler C-N et al (2005) Improving recommendation lists through topic diversification. In: Proceedings of the 14th international conference on World Wide Web, WWW 2005, Chiba, Japan, 10–14 May 2005, pp 22–32

  48. 48.

    Zolaktaf Z, Babanezhad R, Pottinger R (2018) A generic top-N recommendation framework for trading-off accuracy, novelty, and coverage. In: 2018 IEEE 34th international conference on data engineering (ICDE). IEEE, pp 149–160

Download references

Acknowledgements

We thank anonymous reviewers for their very useful comments and suggestions. This work was supported by JSPS KAKENHI (Grant Numbers JP20K01983, JP18H00904, JP17H01001).

Author information

Affiliations

Authors

Corresponding author

Correspondence to Tsukasa Ishigaki.

Additional information

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

Verify currency and authenticity via CrossMark

Cite this article

Lo, K., Ishigaki, T. PPNW: personalized pairwise novelty loss weighting for novel recommendation. Knowl Inf Syst (2021). https://doi.org/10.1007/s10115-021-01546-8

Download citation

Keywords

  • Recommender systems
  • Collaborative filtering
  • Novel recommendation
  • Personalized recommendation
  • Loss weighting