1School of Information Management, Nanjing University, Nanjing 210023, China 2Institute of Government Data Resources,Nanjing University,Nanjing 210023, China 3School of Automation Science and Electrical Engineering, Beihang University, Beijing 100191, China
[Objective] This paper proposes a decomposition model for potential factors based on time series drift, aiming to capture the characteristics of changing user interests and improve the recommendation accuracy. [Methods] First, we built a model combining the temporal dynamic evolution of user preferences and the impacts of their previous behaviors on current ones. Then, we constructed an auxiliary matrix to capture the evolution of users. Finally, we introduced a time impact factor to balance the influence of current and past behaviors. [Results] We examined our model with three experimental datasets. Compared with the baseline method, the accuracy was improved by 40.02%, 3.75% and 19.81% on average. [Limitations] The evolution analysis of interest drift relies on historical data. When the amount of historical data is too sparse, other user information needs to be used for a cold start. [Conclusions] The proposed model has stronger generalization ability to process the characteristics of interest fluctuation, which accurately analyzes user interest evolution, and effectively improves the recommendation performance of enterprises.
丁浩, 胡广伟, 王婷, 索炜. 基于时序漂移的潜在因子模型推荐方法*[J]. 数据分析与知识发现, 2022, 6(10): 1-8.
Ding Hao, Hu Guangwei, Wang Ting, Suo Wei. Recommendation Method for Potential Factor Model Based on Time Series Drift. Data Analysis and Knowledge Discovery, 2022, 6(10): 1-8.
(Ding Hao, Li Shuqing. Personalized Recommendation Based on Predictive Analysis of User’s Interests[J]. Data Analysis and Knowledge Discovery, 2019, 3(11): 43-51.)
(Ding Hao, Ai Wenhua, Hu Guangwei, et al. A Personalized Recommendation Model with Time Series Fluctuation of User Interest[J]. Data Analysis and Knowledge Discovery, 2021, 5(11): 45-58.)
(Zhao Liang, Chen Pinghua, Liao Weiping. Social Recommendation Based on Latent Factors of Social Network Users[J]. Computer Engineering and Applications, 2020, 56(24): 169-174.)
doi: 10.3778/j.issn.1002-8331.1909-0339
(Li Lin, Wang Peipei, Gu Peng, et al. Distributed Singular Value Decomposition Recommendation Algorithm Based on LU Decomposition and Alternating Least Square[J]. Pattern Recognition and Artificial Intelligence, 2020, 33(1): 32-40.)
doi: 10.16451/j.cnki.issn1003-6059.202001004
[5]
Koren Y. Collaborative Filtering with Temporal Dynamics[J]. Communications of the ACM, 2010, 53(4):89-97.
[6]
Zhang C Y, Wang K, Yu H K, et al. Latent Factor Transition for Dynamic Collaborative Filtering[C]// Proceedings of the 2014 SIAM International Conference on Data Mining. Society for Industrial and Applied Mathematics, 2014: 452-460.
[7]
Rafailidis D, Kefalas P, Manolopoulos Y. Preference Dynamics with Multimodal User-Item Interactions in Social Media Recommendation[J]. Expert Systems with Applications, 2017, 74:11-18.
doi: 10.1016/j.eswa.2017.01.005
(Li Danyang, Gan Mingxin. Music Recommendation Method Based on Multi-source Information Fusion[J]. Data Analysis and Knowledge Discovery, 2021, 5(2):94-105.)
[9]
Sun B S, Dong L Y. Dynamic Model Adaptive to User Interest Drift Based on Cluster and Nearest Neighbors[J]. IEEE Access, 2017, 5:1682-1691.
doi: 10.1109/ACCESS.2017.2669243
[10]
Li T Y, Jin L L, Wu Z B, et al. Combined Recommendation Algorithm Based on Improved Similarity and Forgetting Curve[J]. Information, 2019, 10(4). https://doi.org/10.3390/info10040130.
[11]
Luo X, Sun J P, Wang Z D, et al. Symmetric and Nonnegative Latent Factor Models for Undirected, High-Dimensional, and Sparse Networks in Industrial Applications[J]. IEEE Transactions on Industrial Informatics, 2017, 13(6):3098-3107.
doi: 10.1109/TII.2017.2724769
[12]
Chua F C T, Oentaryo R J, Lim E P. Modeling Temporal Adoptions Using Dynamic Matrix Factorization[C]// Proceedings of the13th International Conference on Data Mining. IEEE, 2013.
[13]
Liang X, Quan Y, Zhao S W, et al. Temporal Recommendation on Graphs via Long- and Short-term Preference Fusion[C]// Proceedings of the 16th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. 2010:723-732.
(Zhu Zhenguo, Liu Minkang, Zhao Kaixuan. Recommendation Algorithm Based on Combined Similarity of Users[J]. Computer Systems & Applications, 2018, 27(5): 126-132.)
[15]
Vaca C K, Mantrach A, Jaimes A, et al. A Time-based Collective Factorization for Topic Discovery and Monitoring in News[C]// Proceedings of the International Conference on World Wide Web. 2014:527-538.
[16]
Li K K, Zhou X Z, Lin F, et al. Sparse Online Collaborative Filtering with Dynamic Regularization[J]. Information Sciences, 2019, 505:535-548.
doi: 10.1016/j.ins.2019.07.093
[17]
Wangwatcharakul C, Wongthanavasu S. Dynamic Collaborative Filtering Based on User Preference Drift and Topic Evolution[J]. IEEE Access, 2020, 8: 86433-86447.
doi: 10.1109/ACCESS.2020.2993289
[18]
Lee D D, Seung H S. Learning the Parts of Objects by Non-negative Matrix Factorization[J]. Nature, 1999, 401(675):788-791.
doi: 10.1038/44565
[19]
Lee D D, Seung H S. Algorithms for Non-negative Matrix Factorization[J]. Advances in Neural Information Processing Systems, 2001, 13:556-562.
[20]
McAuley J, Leskovec J. Hidden Factors and Hidden Topics: Understanding Rating Dimensions with Review Text[C]// Proceedings of the 7th ACM Conference on Recommender Systems. ACM, 2013:165-172.
[21]
Wu T, Feng Y, Shang J X, et al. A Novel Recommendation Algorithm Incorporating Temporal Dynamics, Reviews and Item Correlation[J]. IEICE Transactions on Information and Systems, 2018, E101-D (8):2027-2034.
doi: 10.1587/transinf.2017EDP7387