隨著電子商務平台的蓬勃發展，非結構化的評論內容已成為蘊含消費者偏好與需求的關鍵資訊來源。然而，傳統推薦系統多聚焦於明確的使用者-商品互動模式，而忽略評論中豐富的語意資訊，導致在語意複雜且資料稀疏的市場環境中推薦準確性與個人化程度受限，尤以商品種類繁多、長尾特性顯著的益生菌市場為甚。
本研究旨在設計結合主題建模的個人化混合推薦系統架構，針對益生菌市場的消費需求進行精準分析與個人化推薦。應用BERTopic主題建模技術，並結合深度混合推薦系統(Deep Hybrid Collaborative Filtering with content, DHCF)模型。研究選取益生菌市場作為實證場域，其評論內容兼具專業術語與主觀體驗描述的特性。首先從電商平台收集益生菌產品的消費者行為資料和評論文本共103,724筆，進行資料清洗與預處理。接著利用BERTopic提取消費者主題偏好，並將其整合至DHCF模型，進而生成推薦分數。最後，透過RMSE、Precision、Recall、F1 score等指標，評估系統的準確性與效能。
實驗結果顯示，本研究模型在評估指標RMSE上明顯優於DHCF、深度協同過濾以及協同過濾模型，在Precision、Recall及F1 score的表現同樣顯示本研究模型在實際推薦情境上的準確性優於其他對照模型；進一步分析指出，主題建模能串連主題與商品特性之間關聯，使推薦系統能推薦更加個人化產品予消費者，有效命中消費者真實個人保健需求，並可使企業發展差異化行銷策略等重要商業決策。本研究提出的結合主題建模與深度混合推薦架構不僅能顯著提升益生菌商品之個人化推薦效能，亦為功能性食品電商建立可解釋且可擴充的個人化框架，並對企業營運決策具實質價值。

As e-commerce platforms continue to expand, unstructured review content has become a crucial source of insights into consumer preferences and market demands. Traditional recommendation systems, however, predominantly rely on explicit user-item interactions, and often neglecting the nuanced semantic information contained in user reviews. This limitation is particularly prominent in markets with diverse products and pronounced long tail characteristics, such as the probiotics sector, resulting in constrained recommendation accuracy and personalization.
To overcome these limitations, this study proposes a personalized hybrid recommendation system that incorporates topic modeling using BERTopic and a Deep Hybrid Collaborative Filtering with content (DHCF) model. Considering the promise of the probiotics market, this work uses this market as case study to illustrate the feasibility and effectiveness of the proposed system. The proposed system analyzes review topics to capture both product features and consumer needs for enhanced recommendations.
Experimental results of this work demonstrate that the proposed system significantly outperforms DHCF, deep collaborative filtering, and collaborative filtering models in RMSE, precision, recall, and F1 score. Incorporating topic modeling empowers the system to better align product recommendations with individual consumer needs, enhancing the personalization and efficacy of recommendations. These findings highlight the potential of topic modeling-enhanced hybrid recommendation systems to improve personalization, interpretability, and strategic decision-making for e-commerce platforms.

Afoudi, Y., Lazaar, M., & Al Achhab, M. (2021). Hybrid recommendation system combined content-based filtering and collaborative prediction using artificial neural network. Simulation Modelling Practice and Theory, 113, 102375.
Aggarwal, C. C. (2016). Model-based collaborative filtering. In C. C. Aggarwal (Ed.), Recommender systems: The textbook (pp. 71–138). Springer
Alamdari, P. M., Navimipour, N. J., Hosseinzadeh, M., Safaei, A. A., & Darwesh, A. (2020). A Systematic Study on the Recommender Systems in the E-Commerce. IEEE Access, 8, 115694-115716.
Annunziata, A., & Vecchio, R. (2013). Consumer perception of functional foods: A conjoint analysis with probiotics. Food Quality and Preference, 28(1), 348-355.
Ansari, A., Li, Y., & Zhang, J. Z. (2018). Probabilistic topic model for hybrid recommender systems: A stochastic variational Bayesian approach. Marketing Science, 37(6), 987-1008.
Avila, B. P., da Rosa, P. P., Fernandes, T. A., Chesini, R. G., Sedrez, P. A., de Oliveira, A. P. T., Mota, G. N., Gularte, M. A., & Roll, V. F. B. (2020). Analysis of the perception and behaviour of consumers regarding probiotic dairy products. International Dairy Journal, 106, Article 104703.
Badriyah, T., Wijayanto, E. T., Syarif, I., & Kristalina, P. (2017). A hybrid recommendation system for E-commerce based on product description and user profile. In 2017 Seventh International Conference on Innovative Computing Technology (INTECH), pp. 95-100. IEEE.
Bai, L., Hu, M., Ma, Y., & Liu, M. (2019). A hybrid two-phase recommendation for group buying e-commerce applications. Applied Sciences, 9(15), 3141.
Barragáns-Martínez, A. B., Costa-Montenegro, E., Burguillo, J. C., Rey-López, M., Mikic Fonte, F. A., & Peleteiro, A. (2010). A hybrid content-based and item-based collaborative filtering approach to recommend TV programs enhanced with singular value decomposition. Information Sciences, 180(22), 4290-4311.
Blei, D. M. (2009). Topic models. In M. W. Berry & J. Kogan (Eds.), Text mining: Applications and theory (pp. 101–124). Chapman & Hall/CRC.
Blei, D. M., Ng, A. Y., & Jordan, M. I. (2003). Latent dirichlet allocation. Journal of machine Learning research, 3(Jan), 993-1022.
Catherine, R., & Cohen, W. W. (2017). TransNets: Learning to transform for recommendation. In Proceedings of the 11th ACM Conference on Recommender Systems (pp. 288–296). ACM.
Çelik, A. D. (2023). Consumers' perceptions about probiotic food products and their effects on purchase intention: A case study of Eastern Mediterranean Region of Turkey. New Medit, 22(4), 139-154.
Darban, Z. Z., & Valipour, M. H. (2022). GHRS: Graph-based hybrid recommendation system with application to movie recommendation. Expert Systems with Applications, 200, Article 116850.
Darraz, N., Karabila, I., El-Ansari, A., Alami, N., & El Mallahi, M. (2025). Advancing recommendation systems with DeepMF and hybrid sentiment analysis: Deep learning and lexicon-based integration. Expert Systems with Applications, 279, 120987.
Grootendorst, M. (2022). BERTopic: Neural topic modeling with a class-based TF-IDF procedure. arXiv preprint arXiv:2203.05794.
Guia, M., Silva, R. R., & Bernardino, J. (2019). A hybrid ontology-based recommendation system in e-commerce. Algorithms, 12(11), 239.
Hailu, G., Boecker, A., Henson, S., & Cranfield, J. (2009). Consumer valuation of functional foods and nutraceuticals in Canada. A conjoint study using probiotics. Appetite, 52(2), 257-265.
Hamedani, E. M., & Kaedi, M. (2019). Recommending the long tail items through personalized diversification. Knowledge-Based Systems, 164, 348-357.
Haviana, S. F. C., Mulyono, S., Salim, C. A., Adli, M. F., Febriansah, M. T., Sulaiman, N. S., & Yaqob, A. (2024). Leveraging Transformer-Based Topic Modeling using BERTopic for Publication Venue Recommendation. In 2024 11th International Conference on Electrical Engineering, Computer Science and Informatics (EECSI), pp. 556-562. IEEE
He, X., Liao, L., Zhang, H., Nie, L., Hu, X., & Chua, T.-S. (2017). Neural collaborative filtering. In Proceedings of the 26th international conference on world wide web, pp. 173-182. ACM.
Hill, A., Goo, K., & Agarwal, P. (2025). Recommending the right academic programs: An interest mining approach using BERTopic. Data Mining and Knowledge Discovery, 39(3), Article 20.
Huang, X., & Wu, F. (2019). A novel topic-based framework for recommending long-tail products. Computers & Industrial Engineering, 137, 106063
Im, I., & Hars, A. (2007). Does a one-size recommendation system fit all? The effectiveness of collaborative filtering-based recommendation systems across different domains and search modes. ACM Transactions on Information Systems, 26(1), Article 4.
Joachims, T. (1997). A probabilistic analysis of the Rocchio algorithm with TF-IDF for text categorization. In Proceedings of the 14th International Conference on Machine Learning (pp. 143–151). Morgan Kaufmann.
Kawai, M., Sato, H., & Shiohama, T. (2022). Topic model-based recommender systems and their applications to cold-start problems. Expert Systems with Applications, 202, 117129.
Keikhosrokiani, P., & Fye, G. M. (2023). A hybrid recommender system for health supplement e-commerce based on customer data implicit ratings. Multimedia Tools and Applications, 82(15), 22281–22305.
Khan, Z., Iltaf, N., Afzal, H., & Abbas, H. (2020). DST-HRS: A topic driven hybrid recommender system based on deep semantics. Computer Communications, 156, 183-191.
Khan, Z. Y., Niu, Z., Nyamawe, A. S., & ul Haq, I. (2021). A deep hybrid model for recommendation by jointly leveraging ratings, reviews and metadata information. Engineering Applications of Artificial Intelligence, 97, 104066.
Kim, D., Park, C., Oh, J., Lee, S., & Yu, H. (2016). Convolutional matrix factorization for document context-aware recommendation. In Proceedings of the 10th ACM conference on recommender systems, pp. 233-240. ACM.
Ko, H., Lee, S., Park, Y., & Choi, A. (2022). A Survey of Recommendation Systems: Recommendation Models, Techniques, and Application Fields. Electronics, 11(1), 48.
Kordabadi, M., Nazari, A., & Mansoorizadeh, M. (2022). A movie recommender system based on topic modeling using machine learning methods. International Journal of Web Research, 5(2), 19-28.
Koren, Y., Bell, R., & Volinsky, C. (2009). MATRIX FACTORIZATION TECHNIQUES FOR RECOMMENDER SYSTEMS. Computer, 42(8), 30-37.
Lee, J. H., Wood, J., & Kim, J. (2021). Tracing the trends in sustainability and social media research using topic modeling. Sustainability, 13(3), 1269.
Li, L., Chen, J., & Raghunathan, S. (2020). Informative Role of Recommender Systems in Electronic Marketplaces: A Boon or a Bane for Competing Sellers. MIS Quarterly, 44(4), 2123–2154.
Lika, B., Kolomvatsos, K., & Hadjiefthymiades, S. (2014). Facing the cold start problem in recommender systems. Expert Systems with Applications, 41(4), 2065-2073.
Lops, P., de Gemmis, M., & Semeraro, G. (2011). Content-based recommender systems: State of the art and trends. In F. Ricci, L. Rokach, B. Shapira, & P. B. Kantor (Eds.), Recommender systems handbook (pp. 73–105). Springer.
McInnes, L., Healy, J., & Astels, S. (2017). hdbscan: Hierarchical density based clustering. Journal of open source software, 2(11), 205.
McInnes, L., Healy, J., & Melville, J. (2018). Umap: Uniform manifold approximation and projection for dimension reduction. arXiv preprint arXiv:1802.03426.
Nilashi, M., Ibrahim, O., & Bagherifard, K. (2018). A recommender system based on collaborative filtering using ontology and dimensionality reduction techniques. Expert Systems with Applications, 92, 507-520.
Park, S.-H., & Han, S. P. (2012). Empirical analysis of the impact of product diversity on long-term performance of recommender systems. In Proceedings of the 14th Annual International Conference on Electronic Commerce, pp. 280-281. ACM.
Reimers, N. (2019). Sentence-BERT: Sentence Embeddings using Siamese BERT Networks. arXiv preprint arXiv:1908.10084.
Singh, J., Hazela, B., Asthana, P., Singh, S., & Tiwari, A. K. (2023). Recipe Recommender System Using BERTopic Modelling Technique. In Proceedings of the 5th International Conference on Information Management & Machine Intelligence, pp. 1-9. Springer.
Soccol, C. R., Vandenberghe, L. P. D., Spier, M. R., Medeiros, A. B. P., Yamaguishi, C. T., Lindner, J. D., Pandey, A., & Thomaz-Soccol, V. (2010). The Potential of Probiotics: A Review. Food Technology and Biotechnology, 48(4), 413-434.
Son, J., & Kim, S. B. (2017). Content-based filtering for recommendation systems using multiattribute networks. Expert Systems with Applications, 89, 404-412.
Thorat, P. B., Goudar, R. M., & Barve, S. (2015). Survey on collaborative filtering, content-based filtering and hybrid recommendation system. International Journal of Computer Applications, 110(4), 31-36.
Vullam, N., Vellela, S. S., Reddy, V., Rao, M. V., SK, K. B., & Roja, D. (2023). Multi agent personalized recommendation system in e-commerce based on user. In 2023 2nd International Conference on Applied Artificial Intelligence and Computing (ICAAIC), pp. 1194-1199. IEEE.
Wang, C., Ma, Y., Shi, Y. S., & Chen, G. Q. (2024). A type-2 fuzzy review topic-based model for personalized recommendation. Electronic Commerce Research, 23 , 89 116.
Wang, H., Wang, N., & Yeung, D.-Y. (2015). Collaborative deep learning for recommender systems. In Proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining, pp. 1235-1244. ACM.
Wang, Y., Zhu, Y., Zhang, Z., Liu, H., & Guo, P. (2021). Design of hybrid recommendation algorithm in online shopping system. Journal of New Media, 3(4), 119-129.
Williams, N. T. (2010). Probiotics. American Journal of Health-System Pharmacy, 67(6), 449-458.
Wójcik, F., & Górnik, M. (2020). Improvement of E-Commerce Recommendation Systems with Deep Hybrid Collaborative Filtering with Content: A Case Study. Econometrics, 24(3), 37-50.
Zheng, L., Noroozi, V., & Yu, P. S. (2017). Joint deep modeling of users and items using reviews for recommendation. In Proceedings of the tenth ACM international conference on web search and data mining, pp. 425-434. ACM.