Research on the Efficacy of Intelligent Algorithm-Driven Personalized Recommender System in Digital Marketing

Jundong Qian

doi:10.62051/ijgem.v7n3.24

Authors

Jundong Qian

DOI:

https://doi.org/10.62051/ijgem.v7n3.24

Keywords:

Personalized recommendation, Reinforcement learning, Digital marketing, Behavioral modeling

Abstract

In this paper, a personalized recommendation model integrating deep interest network (DNN) and reinforcement learning strategy is constructed to improve the recommendation accuracy and user conversion rate in digital marketing scenarios. The model introduces an attention mechanism on the basis of DNN to weight the user's historical behavior, and dynamically optimizes the recommendation strategy by combining the environmental feedback. In the empirical study on Taobao e-commerce platform, the model outperforms ItemCF and traditional DNN methods in terms of CTR, CVR and user retention, with the click-through rate reaching 9.85%, 59.3% higher than ItemCF, and the conversion rate increasing to 3.18%. The results validate the effectiveness and promotion potential of the model.

Downloads

Download data is not yet available.

References

[1] Khan U H, Naz A, Alarfaj K F, et al. A transformer-based architecture for collaborative filtering modeling in personalized recommender systems. [J]. Scientific reports, 2025, 15 (1): 24503.

[2] Yongsheng Z, Jiaqing L, Ying L, et al. The performance monitoring system for a hydrostatic turntable: an improved intelligent algorithm based on the IPSO-NN model [J]. Industrial Lubrication and Tribology, 2025, 77 (5): 859-869.

[3] Guo Q, Wang F, Cheng S, et al. Fault location and isolation technology for power grid automation based on intelligent algorithms [J]. Energy Informatics, 2025, 8 (1): 88-88.

[4] Stolarz M. Integration of Plant Electrophysiology and Time-Lapse Video Analysis via Artificial Intelligence for the Advancement of Precision Agriculture [J]. Sustainability, 2025, 17 (12): 5614-5614.

[5] Park S, Lee J H, Song H S, et al. An Artificial Intelligence Algorithm for Early Detection of Left Ventricular Systolic Dysfunction in Patients with Normal Sinus Rhythm [J]. Journal of Clinical Medicine, 2025, 14 (12): 4257-4257.

[6] Matthews J M, Su R, Yonish L, et al. A Review of Artificial Intelligence, Algorithms, and Robots Through the Lens of Stakeholder Theory [J]. Journal of Management, 2025, 51 (6): 2627-2676.

[7] Yang L, Yang X. Development of intelligent algorithms for distributed fiber disturbance detection in complex environments [J]. Journal of Computational Methods in Sciences and Engineering, 2025, 25 (4): 3325-3339.

[8] Malik V, Goyal B S. Quantum AI powered dynamic user profiling for next-generation personalized recommender system [J]. Quantum Machine Intelligence, 2025, 7 (1): 44-44.

[9] Starke C, Metikoš L, Helberger N, et al. Contesting personalized recommender systems: a cross-country analysis of user preferences [J]. Information, Communication & Society, 2025, 28 (1): 41-60.

[10] Sharifbaev N A, Zainidinov N H, Kovalev V I, et al. Increasing the Effectiveness of Personalized Recommender Systems Based on the Integrated GNN-RL Model [J]. Journal of Machinery Manufacture and Reliability, 2024, 53 (8): 980-986.

[11] Weikai L. Analyzing the Impact of Information Features on User Continuance Intent in Recommendation Systems [J]. International Journal on Semantic Web and Information Systems (IJSWIS), 2024, 20 (1): 1-36.

[12] Chen R. The application of data mining in data analysis [C]. Proceedings of the International Conference on Mathematics, Modeling, and Computer Science (MMCS2022), 2023, 12625: 473-478. SPIE.

[13] Zhao Z. A Comprehensive Review of Reinforcement Learning in Intelligent Allocation and Optimization of Educational Resources [J]. Journal of Economic Theory and Business Management, 2024, 1(6): 15-24.