In the current context of information overload and increasingly diverse user needs on e-commerce platforms, collaborative filtering and content recommendation algorithms struggle to deeply analyze users' dynamic and fine-grained intent. To this end, this study proposes and implements a personalized recommendation framework that integrates dynamic knowledge graphs with causal inference. First, this paper constructs a dynamic product-attribute-scenario knowledge graph that integrates real-time user behavior sequences. Next, causal inference techniques are introduced, and finally, an exploration-exploitation strategy based on proximal policy optimization (PPO) is discussed. Experimental results show that during a four-week test, the framework achieved an average session duration of 43 minutes for the experimental group. In a bias correction experiment, it more accurately identified and recommended potentially high-quality long-tail products. In an exploration-exploitation balance experiment, the compound reward strategy achieved a 47.2% success rate in exploring new categories in the fourth training cycle. These results demonstrate that the proposed personalized recommendation framework can effectively enhance user stickiness and loyalty, significantly improve the diversity and fairness of the recommendation system, and demonstrate its superiority in discovering users' potential interests.

Li Mu. Data Management Innovation in the e-Commerce Field: Innovation in Personalized Recommendation Algorithms and User Experience Optimization. Academic Journal of Business & Management (2025), Vol. 7, Issue 11: 134-140. https://doi.org/10.25236/AJBM.2025.071118.

[1] Zhang Y. The application of e-commerce recommendation system in smart cities based on big data and cloud computing[J]. Computer Science and Information Systems, 2021, 18(4): 1359-1378.

[2] Gulzar Y, Alwan A A, Abdullah R M, et al. OCA: ordered clustering-based algorithm for e-commerce recommendation system[J]. Sustainability, 2023, 15(4): 2947-2951.

[3] Loukili M, Messaoudi F, El Ghazi M. Machine learning based recommender system for e-commerce[J]. IAES International Journal of Artificial Intelligence, 2023, 12(4): 1803-1811.

[4] Shankar A, Perumal P, Subramanian M, et al. An intelligent recommendation system in e-commerce using ensemble learning[J]. Multimedia Tools and Applications, 2024, 83(16): 48521-48537.

[5] Wu K, Chi K. Enhanced e-commerce customer engagement: A comprehensive three-tiered recommendation system[J]. Journal of Knowledge Learning and Science Technology ISSN: 2959-6386 (online), 2023, 2(3): 348-359.

[6] Wei Fen. Research on design and performance optimization of online recommendation system based on machine learning algorithm[J]. Information and Computer, 2024, 36(3):89-91.

[7] Necula S C, Păvăloaia V D. AI-driven recommendations: A systematic review of the state of the art in e-commerce[J]. Applied Sciences, 2023, 13(9): 5531-5542.

[8] Kim J, Choi I, Li Q. Customer satisfaction of recommender system: Examining accuracy and diversity in several types of recommendation approaches[J]. Sustainability, 2021, 13(11): 6165-6172.

[9] Li W, Cai Y, Hanafiah M H, et al. An empirical study on personalized product recommendation based on cross-border e-commerce customer data analysis[J]. Journal of Organizational and End User Computing (JOEUC), 2024, 36(1): 1-16.

[10] Kim N, Lim H, Li Q, et al. Enhancing Review-Based Recommendations Through Local and Global Feature Fusion[J]. Electronics, 2025, 14(13): 2540-2551.

[11] Liu A, Zhang D, Wang Y, et al. Knowledge graph with machine learning for product design[J]. CIRP Annals, 2022, 71(1): 117-120.

[12] Behera G, Nain N. Collaborative filtering with temporal features for movie recommendation system[J]. Procedia Computer Science, 2023, 218(1): 1366-1373.

[13] Bączkiewicz A, Kizielewicz B, Shekhovtsov A, et al. Methodical aspects of MCDM based E-commerce recommender system[J]. Journal of Theoretical and Applied Electronic Commerce Research, 2021, 16(6): 2192-2229.

[14] Bawack R E, Wamba S F, Carillo K D A, et al. Artificial intelligence in E-Commerce: a bibliometric study and literature review[J]. Electronic markets, 2022, 32(1): 297-338.