Article Details

Abstract

To address the data sparsity issue existing in traditional collaborative filtering recommendation algorithms based on user preference, a recommendation algorithm named RA-UPSDE (Recommendation Algorithm based on User Preferences and Sparse Data Enrichment) is proposed. This algorithm firstly improves the traditional calculation of user preference degrees, then enriches the original rating matrix by considering both the number and the method of data enrichment. Finally, it integrates the user preference similarity with the user rating similarity obtained from the enriched data to calculate and perform collaborative filtering recommendations for the target users. Experimental results demonstrate that the proposed algorithm alleviates the data sparsity issue and further enhances the recommendation quality.

Keywords

user preference; sparse data; similarity; data enrichment

References

1. [1] Li Menghao, Zhao Xuejian, Yu Yunfeng, etc Research progress on recommendation algorithms [J]. Small and Micro Computer Systems, 2022, 43 (3): 544-554
2. [2] Wang Chunbo, Wen Jiwen A review of recommendation research based on heterogeneous information networks [J]. Computer Engineering and Science, 2023, 45 (11): 2047-2059
3. [3] Yu Meng, He Wentao, Zhou Xuchuan, etc Review of Recommendation Systems [J]. Computer Applications, 2022, 42 (6): 1898-1913
4. [4] Wang Shengjie, Zhang Qinghong Privacy clustering based on density weight and recommendation algorithm with improved similarity [J]. Science, Technology and Engineering, 2024, 24 (29): 12623-12630
5. [5] Zhao Xuejian, Li Hao, Tang Haotian Recommendation rating prediction algorithm based on user interest concept grid reduction [J]. Computer Applications, 2023, 43 (11): 3340-3345
6. [6] Li Hao, Liang Jingzhang, Pan Ying An Improved Interest Similarity Personalized Recommendation Algorithm [J]. Computer Technology and Development, 2022, 32 (12): 1-6
7. [7] Ma Xin, Wang Fang Research on Collaborative Filtering Recommendation Algorithm Integrating Category Preference and Data Field Clustering [J]. Modern Intelligence, 2023, 43 (1): 6-18
8. [8] Zhao Junyi, Zhuang Fuzhen, Ao Xiang, etc Overview of Collaborative Filtering Recommendation Systems [J]. Journal of Information Security, 2021, 6 (5): 17-34
9. [9] AHMADIAN S，JOORABLOO N，JALILI M，et al. Alleviating Data Sparsity Problem in Time-Aware Recommender Systems Using a Reliable Rating Profile Enrichment Approach [J/OL].Expert Systems with Applications，2022，187：115849（2021-09-15）.https://doi.org/10.1016/j.eswa.2021.115849.
10. [10] Deng Lele, Huang Jun, Yue Chunlei Collaborative filtering algorithm that integrates project popularity and multiple similarities between users [J]. Small and Micro Computer Systems, 2022, 43 (8): 1631-1637
11. [11] Li Pucong, Wang Shun, Zhong Yuansheng, etc Hybrid recommendation algorithm for material diffusion and heat conduction that integrates trust relationships [J]. Small and Micro Computer Systems, 2021, 42 (10): 2044-2052
12. [12] Zhao Xuejian, Zhang Yuhao, Chen Hao, etc Collaborative filtering recommendation algorithm based on FCM user clustering [J]. Computer Technology and Development, 2021, 31 (8): 6-12

About the authors

Zhang Shaoxuan (1997-), male, Han ethnicity, from Yuncheng, Shanxi Province, is currently pursuing a master's degree in recommendation algorithms.