Personalized career recommendations for university students require models that can capture evolving preferences and support long-term career-development decision-making. However, conventional matching-based approaches often struggle to effectively model temporal dynamics and integrate heterogeneous user information within a unified framework. To address these limitations, this study proposes a hierarchical temporal reinforcement learning framework (HRA-TS) for personalized career recommendation. Specifically, the framework employs a hybrid long short-term memory–graph attention network encoder to jointly capture users’ evolving behavioral evolution and stable personal attributes. A multimodal reward mechanism is introduced to incorporate matching relevance, developmental potential, and behavioral feasibility, thereby guiding policy learning toward more realistic career trajectories. To further enhance training efficiency and decision stability, a curriculum learning strategy is adopted to progressively expand the recommendation space. In addition, federated learning is incorporated to enable privacy-preserving training across distributed clients. Experimental results demonstrate the effectiveness of the proposed framework. Compared with the deep Q-network baseline, HRA-TS improves long-term job matching by 6.02% and cold-start click-through rate by 6.09%. Under the federated learning setting, performance degradation remains below 2.9%, indicating that privacy preservation can be achieved with only limited accuracy loss. These findings suggest that HRA-TS provides an effective and practical solution for personalized career recommendation in dynamic and privacy-sensitive environments.

1. [1] J. T. Iorzua et al., “A machine learning based approach to course and career recommendation system: A systematic literature review,” J. of Computing Theories and Applications, Vol.3, No.1, pp. 1-16. 2025. https://doi.org/10.62411/jcta.12603
2. [2] H. Ding et al. “Capturing dynamic user preferences: A recommendation system model with non-linear forgetting and evolving topics,” Systems, Vol.13, Issue 11, Article No.1034, 2025. https://doi.org/10.3390/systems13111034
3. [3] N. Sritrakool and S. Maneeroj, “Personalized preference drift aware sequential recommender system,” IEEE Access, Vol.9, pp. 155491-155506, 2021. https://doi.org/10.1109/ACCESS.2021.3128769
4. [4] J. Duan et al., “Long short-term enhanced memory for sequential recommendation,” World Wide Web, Vol.26, No.2, pp. 561-583, 2023. https://doi.org/10.1007/s11280-022-01056-9
5. [5] X. Tong, P. Wang, and S. Niu, “Reinforcement learning-based denoising network for sequential recommendation,” Applied Intelligence, Vol.53, No.2, pp. 1324-1335, 2023. https://doi.org/10.1007/s10489-022-03298-6
6. [6] W. Xue et al., “Prefrec: Recommender systems with human preferences for reinforcing long-term user engagement,” Proc. of the 29th ACM SIGKDD Conf. on Knowledge Discovery and Data Mining, pp. 2874-2884, 2023. https://doi.org/10.1145/3580305.3599473
7. [7] L. Wang et al., “Horizontal federated recommender system: A survey,” ACM Computing Surveys, Vol.56, No.9, Article No.240, 2024. https://doi.org/10.1145/3656165
8. [8] R. Gosselin et al., “Privacy and security in federated learning: A survey,” Applied Sciences, Vol.12, Issue 19, Article No.9901, 2022. https://doi.org/10.3390/app12199901
9. [9] C. Sun et al., “Federated deep reinforcement learning for recommendation-enabled edge caching in mobile edge-cloud computing networks,” IEEE J. on Selected Areas in Communications, Vol.41, Issue 3, pp. 690-705, 2023. https://doi.org/10.1109/JSAC.2023.3235443
10. [10] M.-I. Dascălu et al., “Careprofsys – An ontology for career development in engineering designed for the romanian job market,” Revue Roumaine Des Sciences Techniques – Série Électrotechnique Et Énergétique, Vol.68, No.2, pp. 212-217, 2023. https://doi.org/10.59277/RRST-EE.2023.68.2.16
11. [11] B. Vesin, K. Mangaroska, K. Akhuseyinoglu, and M. Giannakos, “Adaptive assessment and content recommendation in online programming courses: On the use of elo-rating,” ACM Trans. Comput. Educ., Vol.22, Issue 3, Article No.33, 2022. https://doi.org/10.1145/3511886
12. [12] V. Ramazanova, M. Sambetbayeva, S. Serikbayeva, Z. Sadirmekova, and A. Yerimbetova, “Development of a Knowledge Graph-Based Model for Recommending MOOCs to Supplement University Educational Programs in Line with Employer Requirements,” IEEE Access, Vol.12, pp. 193313-193331, 2024. https://doi.org/10.1109/ACCESS.2024.3519263
13. [13] S. Ashrafi, B. Majidi, E. Akhtarkavan, and S. H. R. Hajiagha, “Efficient resume-based re-education for career recommendation in rapidly evolving job markets,” IEEE Access, Vol.11, pp. 124350-124367, 2023. https://doi.org/10.1109/ACCESS.2023.3329576
14. [14] M.-I. Dascalu, I.-C. Stanica, I.-A. Bratosin, B.-I. Uta, and C.-N. Bodea, “Virtual reality for career development and exploration: The CareProfSys profiler system case,” Electronics, Vol.13, Issue 13, Article No.2629, 2024. https://doi.org/10.3390/electronics13132629
15. [15] M. Qamhieh, H. Sammaneh, and M. N. Demaidi, “PCRS: Personalized career-path recommender system for engineering students,” IEEE Access, Vol.8, pp. 214039-214049, 2020. https://doi.org/10.1109/ACCESS.2020.3040338
16. [16] E. Padma, P. Sowdharshini, P. Shanmugapriya, K. M. Reshmaa, and C. N. Srimathi. “Career guidance system for students using machine learning,” V. Sharmila, S. Kannadhasan, A. Rajiv Kannan, P. Sivakumar, and V. Vennila (Eds.), “Challenges in Information, Communication and Computing Technology,” pp. 666-671, CRC Press, 2025. https://doi.org/10.1201/9781003559092-115
17. [17] Y. Huang, “Individualized career planning intelligent teaching system for college students based on adaptive artificial intelligence,” J. of Computational Methods in Sciences and Engineering, Vol.25, No.4, 2025. https://doi.org/10.1177/14727978251315731
18. [18] C. Wang, H. Zhu, P. Wang, C. Zhu, X. Zhang, E. Chen, and H. Xiong, “Personalized and explainable employee training course recommendations: A Bayesian variational approach,” ACM Trans. on Information Systems (TOIS), Vol.40, No.4, Article No.70, 2021. https://doi.org/10.1145/3490476
19. [19] S. Maghsudi, A. Lan, J. Xu, and M. van Der Schaar, “Personalized education in the artificial intelligence era: What to expect next,” IEEE Signal Processing Magazine, Vol.38, Issue 3, pp. 37-50, 2021. https://doi.org/37-50.10.1109/MSP.2021.3055032
20. [20] W. Bagunaid, N. Chilamkurti, and P. Veeraraghavan, “AISAR: Artificial intelligence-based student assessment and recommendation system for e-learning in big data,” Sustainability, Vol.14, Issue 17, Article No.10551, 2022. https://doi.org/10.3390/su141710551
21. [21] A. Bhutoria, “Personalized education and artificial intelligence in the United States, China, and India: A systematic review using a human-in-the-loop model,” Computers and Education: Artificial Intelligence, Vol.3, Article No.100068, 2022. https://doi.org/10.1016/j.caeai.2022.100068
22. [22] P. C. Siswipraptini, H. L. H. S. Warnars, A. Ramadhan, and W. Budiharto, “Personalized career-path recommendation model for information technology students in Indonesia,” IEEE Access, Vol.12, pp. 49092-49105, 2024. https://doi.org/10.1109/ACCESS.2024.3381032
23. [23] Y. Ren, “Prediction of career tendency and personalized suggestions for college students under cluster analysis,” J. of Computational Methods in Sciences and Engineering, Vol.25, Issue 2, pp. 1951-1962, 2025. https://doi.org/10.1177/14727978241307474
24. [24] L. Wang, Y. Fu, and Y. Zhang, “A career recommendation method for college students based on occupational values,” Int. J. of Emerging Technologies in Learning, Vol.18, No.1, pp. 201-215, 2023. https://doi.org/10.3991/ijet.v18i01.37127
25. [25] T. V. Yadalam, V. M. Gowda, V. S. Kumar, D. Girish, and Namratha M., “Career recommendation systems using content based filtering,” 2020 5th Int. Conf. on Communication and Electronics Systems (ICCES), pp. 660-665, 2020. https://doi.org/10.1109/ICCES48766.2020.9137992
26. [26] Y. Jiang, “Design of dynamic career path recommendation system based on Markov decision process,” J. of Computational Methods in Sciences and Engineering, Vol.25, Issue 3, pp. 2427-2441, 2025. https://doi.org/10.1177/14727978241313261
27. [27] S. Dhelim, N. Aung, M. A. Bouras, H. Ning, and E. Cambria, “A survey on personality-aware recommendation systems,” Artificial Intelligence Review, Vol.55, pp. 2409-2454, 2022. https://doi.org/10.1007/s10462-021-10063-7
28. [28] J. Chen, H. Dong, X. Wang, F. Feng, M. Wang, and X. He, “Bias and debias in recommender system: A survey and future directions,” ACM Trans. on Information Systems, Vol.41, No.3, Article No.67, 2023. https://doi.org/10.1145/3564284
29. [29] S. Zhang, M. Wang, W. Wang, J. Gao, X. Zhao, Y. Yang et al., “Glint-ru: Gated lightweight intelligent recurrent units for sequential recommender systems,” arXiv preprint, arXiv:2406.10244, 2024. https://doi.org/10.48550/arXiv.2406.10244