P-MOIA-RS: a multi-objective optimization and decision-making algorithm for recommendation systems

Abstract

Besides accuracy, diversity of recommendation list is also important for users. Hence, the optimization of the recommendation system can be abstracted as a multi-objective problem because accuracy and diversity are contradictory goals. Available multi-objective optimization based recommendation schemes return the Pareto set for the target users. However, the scale of Pareto solutions is uncontrollable. If a Pareto set contains too many solutions, it will not be quite useful for users to make the final decision. In this paper, multi-objective immune algorithm is used to improve recommendation accuracy and diversity, then we can get the pareto set. Further, we introduce PROMETHEE into the recommendation system to get a more precise evaluation of Pareto solutions. By combining PROMETHEE with Pareto, we redefine recommendation as Top-n PROMETHEE Pareto optimization problem and a multi-objective immune optimization and decision-making algorithm is presented. The experimental results show that the proposed algorithm, compared with other existing algorithms, can generate more diverse and accurate recommendation list and provide more precise decision-making for the user.

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

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

References

  1. Acilar AM, Arslan A (2009) A collaborative filtering method based on artificial immune network. Expert Syst Appl 36(4):8324–8332

    Article  Google Scholar 

  2. 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 

  3. Aggarwal CC (2016) An introduction to recommender systems. Recommender systems. Springer, Berlin

    Google Scholar 

  4. Bradley K, Smyth B (2001) Improving recommendation diversity. In: Proceedings of the Twelfth Irish conference on Artificial intelligence and cognitive science, Maynooth, Ireland, pp. 85–94

  5. Cacheda F, Formoso V (2011) Comparison of collaborative filtering algorithms: limitations of current techniques and proposals for scalable, high-performance recommender systems. ACM Trans Web 5(1):1–33

    Article  Google Scholar 

  6. Cai Y, Leung HF, Li Q, Min H, Tang J, Li J (2014) Typicality-based collaborative filtering recommendation. IEEE Trans Knowl Data Eng 26(3):766–779

    Article  Google Scholar 

  7. Chakraborty S, Bandhopadyay S (2017) Cotton fibre selection and grading—a PROMETHEE-GAIA-based approach[J]. Int J Cloth Sci Technol 29(5):646–660

    Article  Google Scholar 

  8. Chen L, Zhang G, Zou H, Fast Greedy (2018) MAP inference for determinantal point process to improve recommendation diversity. https://arxiv.org/pdf/1709.05135.pdf. Accessed 26 May 2019

  9. Cui L, Ou P, Fu X, Wen Z, Lu N (2017) A novel multi-objective evolutionary algorithm for recommendation systems. J Parallel Distrib Comput 103:53–63

    Article  Google Scholar 

  10. De Almeida F, Adiel T (2017) Preference modeling experiments with surrogate weighting procedures for the PROMETHEE method[J]. Eur J Oper Res 264(2):453–461

    MathSciNet  MATH  Google Scholar 

  11. Gan M, Jiang R (2013a) Constructing a user similarity network to remove adverse influence of popular objects for personalized recommendation. Expert Syst Appl 40(10):4044–4053

    Article  Google Scholar 

  12. Gan M, Jiang R (2013b) Improving accuracy and diversity of personalized recommendation through power law adjustments of user similarities. Decis Support Syst 55(3):811–821

    Article  Google Scholar 

  13. Geng B, Li L, Jiao L, Gong M, Cai Q, Wu Y (2015) NNIA-RS: a multi-objective optimization based recommender system. Phys A 424:383–397

    MathSciNet  Article  Google Scholar 

  14. Gong M, Jiao L, Du H, Bo L (2008) Multi objective immune algorithm with non-dominated neighbor-based selection. Evolut Comput 16(2):225–255

    Article  Google Scholar 

  15. Hu R, Pu P (2011) Helping users perceive recommendation diversity. Workshop on novelty and diversity in recommender systems (DiveRS 2011), ACM RecSys 2011. Chicago, Illinois, USA

  16. Jiang L, Cheng Y, Yang L, Li J, Yan H, Wang X (2019) A trust-based collaborative filtering algorithm for E-commerce recommendation system. J Ambient Intell Hum Comput 10(8):3023–3034

    Article  Google Scholar 

  17. Kang Z, Peng C, Cheng Q (2016) Top-N recommender system via matrix completion. 30th AAAI conference on Artificial intelligence, AAAI 2016, pp. 179–185. IEEE

  18. Kang Z, Peng C, Cheng Q (2017) Kernel-driven similarity learning. Neurocomputing 267:210–219

    Article  Google Scholar 

  19. Kim J, Lee D, Chung KY (2014) Item recommendation based on context-aware model for personalized u-healthcare service. Multimed Tools Appl 71(2):855–872

    Article  Google Scholar 

  20. Krishankumar R, Ravichandran KS, Saeid AB (2017) A new extension to PROMETHEE under intuitionistic fuzzy environment for solving supplier selection problem with linguistic preferences[J]. Appl Soft Comput J 60(3):564–576

    Google Scholar 

  21. Kunaver M, Požrl T (2017) Diversity in recommender systems—a survey. Knowl-Based Syst 123:154–162

    Article  Google Scholar 

  22. Lacerda A (2017) Multi-objective ranked bandits for recommender systems. Neurocomputing 246:12–24

    Article  Google Scholar 

  23. Linden G, Smith B, York J (2003) Amazon.com recommendations: item-to-item collaborative filtering. IEEE Internet Comput 7(1):76–80

    Article  Google Scholar 

  24. Lü L, Medo M, Chi HY, Zhang YC, Zhang ZK, Zhou T (2012) Recommender systems. Phys Rep 519(1):1–49

    Article  Google Scholar 

  25. Maoguo G, Xiangming J, Hao L (2017) Optimization methods for regularization-based Ill-posed problems: a survey and a multi-objective framework. Front Comput Sci 11(3):362–391

    Article  Google Scholar 

  26. Mikeli A, Apostolou D, Despotis D (2013a) A multi-criteria recommendation method for interval scaled ratings. Ieee/wic/acm international joint conferences on web intelligence (Vol. 3, pp. 9–12). IEEE computer society

  27. Mikeli A, Sotiros D, Apostolou D, Despotis D (2013b) A multi-criteria recommender system incorporating intensity of preferences. International conference on information (Vol. 8255, pp. 1–6). IEEE

  28. Park J (2019) Resource recommender system based on psychological user type indicator. Ambient Intell 10(1):27–39

    Article  Google Scholar 

  29. Ribeiro MT, Lacerda A, Veloso A, Ziviani N (2012) Pareto-efficient hybridization for multi-objective recommender systems. ACM conference on recommender systems (pp. 19–26). ACM

  30. Saldanha WH (2017) Choosing the best evolutionary algorithm to optimize the multi-objective shell-and-tube heat exchanger design problem using PROMETHEE[J]. Appl Therm Eng 127(2):1049–1061

    Article  Google Scholar 

  31. Tomas H, Andre C (2016) evolutionary computing in recommendation systems: a review of recent research[J]. Nat Comput. https://doi.org/10.1007/s11047-016-9540-y

    Article  Google Scholar 

  32. Wang S, Gong M, Li H, Yang J (2016) Multi-objective optimization for long tail recommendation. Knowl Based Syst 104(C):145–155

    Article  Google Scholar 

  33. Xiao J, Wang M, Jiang B, Li J (2018) A personalized recommendation system with combinational algorithm for online learning. Ambient Intell 9(3):667–677

    Article  Google Scholar 

  34. Yang X, Guo Y, Liu Y, Steck H (2014) A survey of collaborative filtering based social recommender systems. Comput Commun 41(5):1–10

    Article  Google Scholar 

  35. Yeung CH (2016) Do recommender systems benefit users? A modeling approach. J Stat Mech Theory Exp 2016(4):043401

    MathSciNet  Article  Google Scholar 

  36. Zhao X et al (2016) Improving diversity of user-based two-step recommendation algorithm with popularity normalization. In: Gao H et al (eds) International conference on database systems for advanced applications. DASFAA 2016 Workshops, LNCS 9645, pp 15–26. https://doi.org/10.1007/978-3-319-32055-7_2

  37. ZhengYi C, Xue-yang Y, Ya-lun L (2016) Throughput optimization in cognitive wireless network based on clone selection algorithm. Comput Electr Eng 52(1):328–336

    Article  Google Scholar 

  38. Zuo Y, Gong M, Zeng J, Ma L, Jiao L (2015) Personalized recommendation based on evolutionary multi-objective optimization [research frontier]. IEEE Comput Intell Mag 10(1):52–62

    Article  Google Scholar 

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant Nos. 61972456, 41772123); Natural Science Foundation of Tianjin (No. 19JCYBJC15800).

Author information

Affiliations

Authors

Corresponding author

Correspondence to Yalun Li.

Ethics declarations

Conflict of interest

We declare that we have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants performed by any of the authors.

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

Chai, Z., Li, Y. & Zhu, S. P-MOIA-RS: a multi-objective optimization and decision-making algorithm for recommendation systems. J Ambient Intell Human Comput 12, 443–454 (2021). https://doi.org/10.1007/s12652-020-01997-x

Download citation

Keywords

  • Recommendation accuracy
  • Recommendation diversity
  • Multi-objective immune optimization
  • PARETO Refinement
  • PROMETHEE
  • Recommendation system