Article Details

Abstract

Aiming at the problem of attribute redundancy or irrelevant attribute feature interference in traditional content-based recommendation algorithms, a Literature Recommendation algorithm based on Reliable User Information (LR_RUI) is proposed. It improves the traditional user portrait portrayal, introduces the time decay factor, dynamically adjusts the user interest features, and determines the number of attributes required through the proposed probability model. It selects the key attributes to obtain the reliable user information, and finally generates a literature recommendation list more in line with the personalised preference of the users based on the similarity among the users. The experimental results show that compared with the traditional algorithm CB, NDCG@10 improves by 0.11, Precision improves by 0.2, and MRR improves by 0.3, which better achieves personalised recommendation.

Keywords

recommendation algorithm; user preference; reliable user information; literature recommendation

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] Hu Qi, Zhu Dingju, Wu Huixi, etc A Review of Research on Intelligent Recommendation Systems [J]. Computer System Applications, 2022, 31 (4): 47-58
3. [3] LIAN Z，HUI C. Research on Recommendation Algorithms in Web of Things System [C]//2014 7th International Conference on Intelligent Computation Technology and Automation.Changsha：IEEE，2014：569-572.
4. [4] SUN X, XU X L, XIA F. CROA: A Content-Based Recommendation Optimization Algorithm for Personalized Knowledge Services [J].2019 IEEE 21st International Conference on High Performance Computing and Communications，Zhangjiajie：IEEE，2019：804-810.
5. [5] Diao Yu, Xue Hong Research on TF-IDF Literature Recommendation Method Based on Electronic Resource Behavior Data - Taking Electronic Resource Off campus Access System as an Example [J]. Library Journal, 2022, 41 (12): 45-54
6. [6] CHAUDHURI A，SARMA M，SAMANTA D. SHARE: Designing Multiple Criteria-based Personalized Research Paper Recommendation System [J].Information Sciences，2022，617：41-64.
7. [7] Hou Meng, Wang Guopeng, Song Lizhe, etc Research on Collaborative Filtering Recommendation Algorithm Based on Improved User Profile [J]. Research on Printing and Digital Media Technology, 2024 (6): 117-123+134
8. [8] 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.
9. [9] Chi Yuzhuo, Li Jincan Academic Paper Recommendation Model Based on Multidimensional Metric Weights [J]. Shandong Library Journal, 2024 (5): 39-48
10. [10] Zhang Xiaojuan, Liu Yijun, Liu Jie, etc A Review of Research on Personalized Chemistry Paper Recommendations [J]. Journal of Information Science, 2024, 43 (1): 106-126
11. [11] Wang Meng Research on Document Reordering Model Integrating Topic Relevance and Search Conversations [D]. Yinchuan: Northern University for Nationalities, 2024
12. [12] Chen Zhen, Xie Denghui, Wang Xiaolong, etc A high-order complementary cloud API recommendation method based on probabilistic logical reasoning [J]. Journal of Computer Science, 2024, 47 (8): 1922-1948
13. [13] Zhu Jun, Han Lixin, Zong Ping, etc Adaptive Location Recommendation Algorithm Based on Time Clustering and User Dynamic Similarity [J]. Computer Science, 2023, 50 (12): 113-122

About the authors

Li Jiahui (2000-), female, Han ethnicity, from Lvliang, Shanxi, currently pursuing a master's degree in recommendation algorithms.