由於資訊科技發展迅速，近年來推薦系統在各領域已逐漸被重視，其主要目的為將使用者可能喜好的項目推薦給使用者。然而使用者喜好的商品並不會一成不變，而是會隨著時間推移出現變化，此問題被稱為偏好漂移(Preference Drifting)，若推薦系統所推薦之商品未隨著使用者的偏好漂移出現變化，便會降低推薦準確性，使用者也將不再信任該推薦系統。
協同過濾推薦方法(Collaborative Filtering Approach)為近年來較為受歡迎的推薦方法，其中Lo et al. (2018)所提出之時間矩陣分解法(Temporal Matrix Factorization, TMF)為偏好漂移問題中主要的研究方法，且已被證明能夠學習出更準確的使用者隱含特徵與商品隱含特徵。該方法將使用者偏好漂移假設為線性關係，然而使用者偏好會被許多因素所影響，因此假設為線性關係較為不合理。
為了解決此問題，本研究基於TMF之架構，將其捕捉使用者偏好漂移之線性回歸系統改良為非線性函數中的對數函數，並搭配主題模型分析使用者所撰寫之商品評論，將其結果用於初始化矩陣分解，以提升模型穩定度及推薦的準確性。
於實驗結果中，本研究使用Amazon Review Data資料集進行訓練。根據結果顯示，本研究模型輸出預測評分與TMF及SVD矩陣分解模型相比，有較為準確之預測結果。

Recommendation systems have gradually been paid attention to in various fields in recent years. The primary purpose is to recommend items that users may like. However, the user prefers items that do not stay the same but change over time. The problem is known as preference drifting. The Temporal Matrix Factorization(TMF) approach proposed by Lo et al. (2018) is the primary research approach to the preference drifting problem. It proves to learn more accurate user and item features. This approach assumes user preference drift as a linear relationship.However, users’ preferences are affected by many factors, so it is unreasonable to consider a linear relationship.
We proposed a model based on the TMF model and improved its linear regression system to the logarithmic function. The topic model analyzes the user’s reviews and uses the results to initialize matrix factorization to improve model stability and recommended accuracy.

Adomavicius, G. and Tuzhilin, A. (2005). Toward the next generation of recommender
systems: a survey of the state-of-the-art and possible extensions. IEEE Transactions
on Knowledge and Data Engineering, 17(6):734–749.

Aizawa, A. (2003). An information-theoretic perspective of tf–idf measures. Information Processing Management, 39(1):45–65.

Beel, J., Gipp, B., Langer, S., and Breitinger, C. (2016). Paper recommender systems:
a literature survey. International Journal on Digital Libraries, 17(4):305–338.

Belford, M., Mac Namee, B., and Greene, D. (2018). Stability of topic modeling via
matrix factorization. Expert Systems with Applications, 91:159–169.

Bobadilla, J., Ortega, F., Hernando, A., and Guti´errez, A. (2013). Recommender systems survey. Knowledge-Based Systems, 46:109–132.

Campos, R., dos Santos, R. P., and Oliveira, J. (2020). A recommendation system based
on knowledge gap identification in moocs ecosystems. In XVI Brazilian Symposium
on Information Systems. Association for Computing Machinery.

Chen, L., Chen, G., andWang, F. (2015). Recommender systems based on user reviews:
the state of the art. User Modeling and User-Adapted Interaction, 25(2):99–154.

Chin, W.-S., Yuan, B.-W., Yang, M.-Y., Zhuang, Y., Juan, Y.-C., and Lin, C.-J. (2016).
Libmf: A library for parallel matrix factorization in shared-memory systems. J.
Mach. Learn. Res., 17(1):2971–2975.

Dey, A. K., Abowd, G. D., and Salber, D. (2001). A conceptual framework and
a toolkit for supporting the rapid prototyping of context-aware applications. Human–Computer Interaction, 16(2-4):97–166.

El-Dosuky, M., Rashad, M. Z., Hamza, T., and El-Bassiouny, A. (2012). Food recommendation using ontology and heuristics. In International conference on advanced machine learning technologies and applications, pages 423–429.

Feng, L., Zhao, Q., and Zhou, C. (2020). Improving performances of top-n recommendations with co-clustering method. Expert Systems with Applications, 143:113078.

Funk, S. (2006). Netflix update: Try this at home.

Huang, Z. and Stakhiyevich, P. (2021). A Time-Aware Hybrid Approach for Intelligent Recommendation Systems for Individual and Group Users. Complexity, 2021.

Ifada, N., Alim, D. R. M., and Sophan, M. K. (2019). Nmf-based dcg optimization
for collaborative ranking on recommendation systems. In Proceedings of the 2019
2nd International Conference on Machine Learning and Machine Intelligence, page
7–11. Association for Computing Machinery.

Kefalas, P. and Manolopoulos, Y. (2017). A time-aware spatio-textual recommender
system. Expert Systems with Applications, 78:396–406.

Khan, Z., Iltaf, N., Afzal, H., and Abbas, H. (2020). Enriching non-negative matrix factorization with contextual embeddings for recommender systems. Neurocomputing, 380:246–258.

Kim, D., Park, C., Oh, J., Lee, S., and Yu, H. (2016). Convolutional matrix factorization
for document context-aware recommendation. In Proceedings of the 10th ACM
conference on recommender systems, pages 233–240.

Koren, Y. (2008). Factorization meets the neighborhood: a multifaceted collaborative filtering model. In Proceedings of the 14th ACM SIGKDD international conference on Knowledge discovery and data mining, pages 426–434.

Koren, Y. (2009). Collaborative filtering with temporal dynamics. Proceedings of the
ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,
pages 447–455.

Koren, Y., Bell, R., and Volinsky, C. (2009). Matrix Factorization Techniques For
Recommender Systems. pages 30–37.

Lai, C.-H., Liu, D.-R., and Lin, S.-R. (2018). Document recommendation with implicit
feedback based on matrix factorization and topic model. In 2018 IEEE International
Conference on Applied System Invention (ICASI), pages 62–65.

Lee, D. D. and Seung, H. S. (1999). Learning the parts of objects by non-negative
matrix factorization. Nature, 401(6755):788–791.

Liu, D.-R., Chen, K.-Y., Chou, Y.-C., and Lee, J.-H. (2018). Online recommendations
based on dynamic adjustment of recommendation lists. Knowledge-Based Systems,161:375–389.

Liu, Y., Liu, Y., Shen, Y., and Li, K. (2017). Recommendation in a changing world:
Exploiting temporal dynamics in ratings and reviews. ACM Trans. Web, 12(1).

Lo, Y. Y., Liao,W., Chang, C. S., and Lee, Y. C. (2018). Temporal Matrix Factorization
for Tracking Concept Drift in Individual User Preferences. IEEE Transactions on
Computational Social Systems, 5(1):156–168.

Lu, J., Wu, D., Mao, M., Wang, W., and Zhang, G. (2015). Recommender system
application developments: a survey. Decision Support Systems, 74:12–32.

Lu, M. and Tian, P. (2018). Matrix factorization recommendation algorithm incorporating tag factor. In 2018 IEEE 4th Information Technology and Mechatronics Engineering Conference (ITOEC), pages 403–407.

Lu, Z., Agarwal, D., and Dhillon, I. S. (2009). A spatio-temporal approach to collaborative filtering. In Proceedings of the third ACM conference on Recommender systems, pages 13–20.

Papadakis, H., Panagiotakis, C., and Fragopoulou, P. (2017). Scor: A synthetic coordinate based recommender system. Expert Systems with Applications, 79:8–19.

Pe˜na, F. J., O’Reilly-Morgan, D., Tragos, E. Z., Hurley, N., Duriakova, E., Smyth, B.,
and Lawlor, A. (2020). Combining Rating and Review Data by Initializing Latent
Factor Models with Topic Models for Top-N Recommendation, page 438–443.

Poirson, E. and Cunha, C. D. (2019). A recommender approach based on customer
emotions. Expert Systems with Applications, 122:281–288.

Rendle, S. (2010). Factorization machines. In 2010 IEEE International conference on
data mining, pages 995–1000. IEEE.

Rendle, S., Freudenthaler, C., Gantner, Z., and Schmidt-Thieme, L. (2012). Bpr:
Bayesian personalized ranking from implicit feedback.

Ricci, F., Rokach, L., and Shapira, B. (2011). Introduction to recommender systems
handbook. In Recommender systems handbook, pages 1–35.

Roy, D., Chowdhury, R. R., Nasser, A. B., Azmi, A., and Babaeianjelodar, M. (2022).
Item recommendation using user feedback data and item profile. arXiv preprint
arXiv:2206.14133.

Salakhutdinov, R. and Mnih, A. (2009). Probabilistic matrix factorization. Advances
in Neural Information Processing Systems 20 - Proceedings of the 2007 Conference, pages 1–8.

Sarwar, B., Karypis, G., Konstan, J., and Riedl, J. (2001). Item-based collaborative
filtering recommendation algorithms. In Proceedings of the 10th international conference on World Wide Web, pages 285–295.

Schafer, B., J, B., Frankowski, D., Dan, Herlocker, Jon, Shilad, and Sen, S. (2007).
Collaborative filtering recommender systems.

Shi, T., Kang, K., Choo, J., and Reddy, C. K. (2018). Short-text topic modeling via
non-negative matrix factorization enriched with local word-context correlations. In
Proceedings of the 2018 World Wide Web Conference, page 1105–1114. International World Wide Web Conferences Steering Committee.

Su, X. and Khoshgoftaar, T. M. (2009). A survey of collaborative filtering techniques.
Advances in artificial intelligence, 2009.

Sun, J. Z., Parthasarathy, D., and Varshney, K. R. (2014). Collaborative kalman filtering for dynamic matrix factorization. IEEE Transactions on Signal Processing,
62(14):3499–3509.

Wangwatcharakul, C. and Wongthanavasu, S. (2021). A novel temporal recommender system based on multiple transitions in user preference drift and topic review evolution. Expert Systems with Applications, page 115626.

Wooldridge, J. M. (2015). Introductory econometrics: A modern approach. Cengage
learning.

Xia, P., Jiang,W.,Wu, J., Xiao, S., andWang, G. (2021). Exploiting temporal dynamics
in product reviews for dynamic sentiment prediction at the aspect level. ACM Trans.
Knowl. Discov. Data, 15(4).

Yan, Y., Yu, G., and Yan, X. (2020). Online doctor recommendation with convolutional neural network and sparse inputs. Computational Intelligence and Neuroscience, 2020.

Zafari, F., Moser, I., and Baarslag, T. (2019). Modelling and analysis of temporal
preference drifts using a component-based factorised latent approach. Expert Systems with Applications, 116:186–208.