Skip to main content

PU Active Learning for Recommender Systems


In recommender systems, supervised information is usually obtained from the historical data of users. For example, if a user watched a movie, then the user-movie pair will be marked as positive. On the other hand, the user-movie pairs did not appear in the historical data could be either positive or negative. This phenomenon motivates us to formalize the recommender task as a Positive Unlabeled learning problem. As the model trained on the biased historical data may not generalize well on future data, we propose an active learning approach to improve the model by querying further labels from the unlabeled data pool. With the target of querying as few instances as possible, an active selection strategy is proposed to minimize the expected loss and match the distribution between labeled and unlabeled data. Experiments are performed on both classification datasets and movie recommendation dataset. Results demonstrate that the proposed approach can significantly reduce the labeling cost while achieving superior performance regarding multiple criteria.

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

Fig. 1
Fig. 2
Fig. 3


  1. Bezdek JC, Hathaway RJ (2003) Convergence of alternating optimization. Neural Parallel Sci Comput 11(4):351–368

    MathSciNet  MATH  Google Scholar 

  2. Cai JJ, Tang J, Chen QG, Hu Y, Wang X, Huang SJ (2019) Multi-view active learning for video recommendation. In: IJCAI, pp 2053–2059

  3. Chang CC, Lin CJ (2011) Libsvm: a library for support vector machines. ACM Trans Intell Syst Technol (TIST) 2(3):1–27

    Article  Google Scholar 

  4. Denis F (1998) Pac learning from positive statistical queries. In: International Conference on Algorithmic Learning Theory, Springer, pp 112–126

  5. Du Plessis M, Niu G, Sugiyama M (2015) Convex formulation for learning from positive and unlabeled data. In: International conference on machine learning, pp 1386–1394

  6. Du Plessis MC, Niu G, Sugiyama M (2014) Analysis of learning from positive and unlabeled data. In: Advances in neural information processing systems, pp 703–711

  7. Elkan C, Noto K (2008) Learning classifiers from only positive and unlabeled data. In: Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, pp 213–220

  8. Floudas CA, Visweswaran V (1995) Quadratic optimization. In: Handbook of global optimization, Springer, pp 217–269

  9. Ghasemi A, Rabiee HR, Fadaee M, Manzuri MT, Rohban MH (2011) Active learning from positive and unlabeled data. In: 2011 IEEE 11th International Conference on Data Mining Workshops, IEEE, pp 244–250

  10. He G, Duan Y, Li Y, Qian T, He J, Jia X (2015) Active learning for multivariate time series classification with positive unlabeled data. In: 2015 IEEE 27th International Conference on Tools with Artificial Intelligence (ICTAI), IEEE, pp 178–185

  11. Hsieh CJ, Natarajan N, Dhillon I (2015) Pu learning for matrix completion. In: International Conference on Machine Learning, pp 2445–2453

  12. Huang SJ, Chen S (2016) Transfer learning with active queries from source domain. In: IJCAI, pp 1592–1598

  13. Huang SJ, Zhou ZH (2013) Active query driven by uncertainty and diversity for incremental multi-label learning. In: 2013 IEEE 13th International Conference on Data Mining, IEEE, pp 1079–1084

  14. Huang SJ, Chen S, Zhou ZH (2015) Multi-label active learning: Query type matters. In: Twenty-Fourth International Joint Conference on Artificial Intelligence

  15. Kelly D, Teevan J (2003) Implicit feedback for inferring user preference: a bibliography. ACM Sigir Forum 37(2):18–28

  16. Kim Y (2014) Convolutional neural networks for sentence classification. arXiv preprint arXiv:1408.5882

  17. Kiryo R, Niu G, Du Plessis MC, Sugiyama M (2017) Positive-unlabeled learning with non-negative risk estimator. In: Advances in neural information processing systems, pp 1675–1685

  18. Li X, Liu B (2003) Learning to classify texts using positive and unlabeled data. IJCAI 3:587–592

    Google Scholar 

  19. Liu B, Lee WS, Yu PS, Li X (2002) Partially supervised classification of text documents. ICML 2:387–394

    Google Scholar 

  20. Niu G, du Plessis MC, Sakai T, Ma Y, Sugiyama M (2016) Theoretical comparisons of positive-unlabeled learning against positive-negative learning. In: Advances in neural information processing systems, pp 1199–1207

  21. Pan R, Zhou Y, Cao B, Liu NN, Lukose R, Scholz M, Yang Q (2008) One-class collaborative filtering. In: 2008 Eighth IEEE International Conference on Data Mining, IEEE, pp 502–511

  22. Settles B (2009) Active learning literature survey. Tech. rep., University of Wisconsin-Madison Department of Computer Sciences

  23. Steck H (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, pp 713–722

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Yuan Jiang.

Ethics declarations

Conflict of interest

COI Affiliations: Nanjing University ( Nanjing University of Aeronautics and Astronautics ( There is no other conflict of interest.

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

Chen, JL., Cai, JJ., Jiang, Y. et al. PU Active Learning for Recommender Systems. Neural Process Lett 53, 3639–3652 (2021).

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • PU learning
  • Active learning
  • Recommender systems
  • Implicit feedback