Abstract
AbstractIn the process of group recommendation, due to the different preferences of group members, the recommendation results cannot meet the needs of all users. How to maximize the fairness of group recommendation is still a challenge. Therefore, this paper proposes a group recommendation algorithm based on user activity. Firstly, a group discovery algorithm based on item cluster preference was used to mine potential groups. Secondly, considering the dynamic change of activity, a sliding time window is designed to investigate the recent activity of each member in the group at the time of subgroup division, and the group is divided into active subgroup and inactive subgroup. Finally, the group recommendation list was generated by aggregating the subgroup preferences by average consensus. Experimental results on the public dataset show that compared with the AGREE algorithm, the recommendation accuracy and coverage of the proposed algorithm are improved by 2.1% and 2.9%, respectively. By focusing on the preference needs of inactive users, the proposed algorithm effectively improves the recommendation satisfaction and group fairness.
Funder
Natural Science Foundation of Gansu Province
Publisher
Springer Science and Business Media LLC
Reference61 articles.
1. Shrivastava, R., Sisodia, D.S., Nagwani, N.K.: Deep neural network-based multi-stakeholder recommendation system exploiting multi-criteria ratings for preference learning. Expert Syst. Appl. 213, 119071 (2023)
2. Breese, J. S., Heckerman, D., Kadie, C.: Empirical analysis of predictive algorithms for collaborative filtering. arXiv preprint arXiv.1301.7363 (2013).
3. Procof the 15th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 447–456 (2009).
4. Ma H, Yang H, Lyu M R, et al.: Sorec: social recommendation using probabilistic matrix factorization. Proceedings of the 17th ACM conference on Information and knowledge management. Napa Valley, USA: ACM. 931–940 (2008).
5. Jia, J., Yao, Y., Chen, W.: Group recommendation algorithm based on non-negative matrix decomposition. Comput. Eng. Sci. 44(05), 933–943 (2022)