基於大型語言模型之對話式推薦系統：以中藥方推薦為例

、線上人數：13

、訪客IP：3.144.176.149

姓名	張廷睿(Ting-Jui Chang) 查詢紙本館藏	畢業系所	資訊工程學系
論文名稱	基於大型語言模型之對話式推薦系統：以中藥方推薦為例 (Conversational Recommender System based on Large Language Model: A Case Study in Traditional Chinese Medicine Recommendation)
檔案	[Endnote RIS 格式] [Bibtex 格式] [相關文章] [文章引用] [完整記錄] [館藏目錄] 至系統瀏覽論文 (2026-12-31以後開放)
摘要(中)	傳統的推薦系統主要依賴於分析數據和機器學習算法，並由系統單方面向使用者推播。對話式推薦系統則可以直接接受來自使用者主動提供的資訊，而系統也能透過文字對項目進行推薦，給予最直接的幫助。在對話式推薦系統（Conversational Recommender System, CRS）中使用大型語言模型（Large Language Model, LLM）能獲得許多傳統模型無法擁有的優勢。首先，LLM的系統不需要經過訓練即可展現出色的性能，能夠解決冷啟動問題。其次，LLM的泛用性及可擴展性極高，能適應或導入到各種應用場景。大多數以往的研究偏向於從大量的物件中推薦出一個最相關的內容，這較適合使用檢索增強生成（Retrieval Augmented Generation, RAG）技術；我們的研究著重於從少量的物件中推薦一個最適合的內容，並強調應依據病因來推薦藥方。本研究首次嘗試以失眠患者的中藥方推薦作為任務目標，使用繁體中文能力出色的gemini-1.5-flash模擬並生成病患與中醫師之間的對話。我們提出Hint Module來導入生理量測及中醫問診技巧，透過偵測特定字串來觸發系統機制，給予LLM額外的提示訊息藉此影響它的輸出結果。我們的實驗結果顯示，此方法可生成寫實的自述，能被視作為良好的對話範例，並在十種中藥方的推薦任務中，可得到八成以上的準確率及Macro-F1成績。其中Hint Module能顯著地改善多輪對話後的成績表現 (p-value < 0.01)。最後我們也以中醫學的各項觀點進行分析，透過視覺化的圖表呈現出各個藥方之間分布上的關聯性，以得到更全面及清晰的瞭解。實驗結果展示了以LLM打造的中藥方推薦系統擁有傑出的基礎能力，並擁有絕佳的可擴展性。
摘要(英)	Traditional recommender systems primarily rely on data analysis and machine learning algorithms, with recommendations being pushed to users unilaterally by the system. In contrast, conversational recommender systems can directly accept information actively provided by users, and the system can recommend items through text, offering the most direct assistance. Using a Large Language Model (LLM) in a Conversational Recommender System (CRS) offers many advantages that traditional models do not have. First, an LLM system does not require training to perform well, which solves the cold-start problem. Second, LLMs possess high versatility and scalability, allowing them to adapt or be integrated into various application scenarios. Most previous studies focused on recommending the most relevant content from a large set of items, which is more suitable for Retrieval Augmented Generation (RAG) techniques. Our research, however, focuses on recommending the most appropriate content from a small set of items, emphasizing the recommendation of prescriptions based on the cause of illness. Our study is the first to attempt using Traditional Chinese Medicine (TCM) prescription recommendations for insomnia patients as a task goal, utilizing gemini-1.5-flash, which excels in Traditional Chinese, to simulate and generate conversations between patients and TCM doctors. We propose using a Hint Module to incorporate physiological measurements and consultation techniques, triggering the system to provide LLM with additional prompt messages by detecting specific strings, thereby influencing the LLM’s output. Our experimental results show that this method can generate realistic patient statements that can be regarded as good examples, achieving over 80% accuracy and Macro-F1 score in recommending ten classes of prescriptions. The Hint Module can significantly (p-value < 0.01) improve the performance of the multiple rounds of dialogue. Lastly, we conducted an analysis based on various perspectives of TCM, we used visualized charts to present the distributional relationships among different prescriptions, aiming for a more comprehensive and clearer understanding. The experimental results show that the TCM prescription recommender system built with LLM has excellent basic capabilities and scalability.
關鍵字(中)	★ 對話式推薦系統 ★ 中藥方推薦 ★ 大型語言模型	關鍵字(英)	★ Conversational Recommender System ★ CRS ★ Traditional Chinese Medicine Recommendation ★ TCM Prescription Recommendation ★ Large Language Model ★ LLM
論文目次	中文摘要 i Abstract ii 誌謝 iv Introduction 1 Related Work 4 2.1 Conversational Recommender System (CRS) 4 2.2 Evaluation Approaches 6 2.3 Related Models & Frameworks 8 2.4 Related Task 10 2.5 TCM Prescription Recommendation 11 Experiment 13 3.1 Experiment Settings 13 3.2 TCM Data Preprocessing 17 3.3 Personas Generation 18 3.4 Conversation Generation 20 Analysis 24 4.1 Direct prediction 24 4.2 Main results 25 4.3 Number of Rounds 28 4.4 Case Study 29 4.5 TCM domain knowledge 31 4.6 Limitation 35 Conclusion and Future work 37 5.1 Conclusion 37 5.2 Future work 38 Bibliography 40 Appendix 44
參考文獻	[1] Y. Zhang, X. Chen, Q. Ai, L. Yang, and W. B. Croft, “Towards Conversational Search and Recommendation: System Ask, User Respond,” in Proceedings of the 27th ACM International Conference on Information and Knowledge Management, Torino Italy: ACM, Oct. 2018, pp. 177–186. doi: 10.1145/3269206.3271776. [2] H. Zhou, M. Huang, Y. Liu, W. Chen, and X. Zhu, “EARL: Informative Knowledge-Grounded Conversation Generation with Entity-Agnostic Representation Learning,” in Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, M.-F. Moens, X. Huang, L. Specia, and S. W. Yih, Eds., Online and Punta Cana, Dominican Republic: Association for Computational Linguistics, Jan. 2021, pp. 2383–2395. doi: 10.18653/v1/2021.emnlp-main.184. [3] V. Sadiri Javadi, M. Potthast, and L. Flek, “OpinionConv: Conversational Product Search with Grounded Opinions,” in Proceedings of the 24th Annual Meeting of the Special Interest Group on Discourse and Dialogue, S. Stoyanchev, S. Joty, D. Schlangen, O. Dusek, C. Kennington, and M. Alikhani, Eds., Prague, Czechia: Association for Computational Linguistics, Sep. 2023, pp. 66–76. doi: 10.18653/v1/2023.sigdial-1.6. [4] Y. Feng et al., “A Large Language Model Enhanced Conversational Recommender System,” Aug. 11, 2023, arXiv: arXiv:2308.06212. Accessed: Jun. 06, 2024. [Online]. Available: http://arxiv.org/abs/2308.06212 [5] R. Li, S. Kahou, H. Schulz, V. Michalski, L. Charlin, and C. Pal, “Towards Deep Conversational Recommendations,” Mar. 04, 2019, arXiv: arXiv:1812.07617. Accessed: Jun. 06, 2024. [Online]. Available: http://arxiv.org/abs/1812.07617 [6] L. Murakhovs’ka, P. Laban, T. Xie, C. Xiong, and C.-S. Wu, “Salespeople vs SalesBot: Exploring the Role of Educational Value in Conversational Recommender Systems,” in Findings of the Association for Computational Linguistics: EMNLP 2023, H. Bouamor, J. Pino, and K. Bali, Eds., Singapore: Association for Computational Linguistics, Feb. 2023, pp. 9823–9838. doi: 10.18653/v1/2023.findings-emnlp.657. [7] X. Wang, X. Tang, X. Zhao, J. Wang, and J.-R. Wen, “Rethinking the Evaluation for Conversational Recommendation in the Era of Large Language Models,” in Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, H. Bouamor, J. Pino, and K. Bali, Eds., Singapore: Association for Computational Linguistics, Feb. 2023, pp. 10052–10065. doi: 10.18653/v1/2023.emnlp-main.621. [8] T.-J. Chang, L. H.-M. Lin, and R. T.-H. Tsai, “Conversational Product Recommendation using LLM,” in 2024 IEEE 4th International Conference on Electronic Communications, Internet of Things and Big Data (ICEIB), Apr. 2024, pp. 340–343. doi: 10.1109/ICEIB61477.2024.10602608. [9] H. Zhang et al., “HuatuoGPT, Towards Taming Language Model to Be a Doctor,” in Findings of the Association for Computational Linguistics: EMNLP 2023, H. Bouamor, J. Pino, and K. Bali, Eds., Singapore: Association for Computational Linguistics, Feb. 2023, pp. 10859–10885. doi: 10.18653/v1/2023.findings-emnlp.725. [10] H. Zhang, X. Wang, Z. Meng, Y. Jia, and D. Xu, “Qibo: A Large Language Model for Traditional Chinese Medicine,” Mar. 24, 2024, arXiv: arXiv:2403.16056. Accessed: Jun. 07, 2024. [Online]. Available: http://arxiv.org/abs/2403.16056 [11] Y. Tan et al., “MedChatZH: A tuning LLM for traditional Chinese medicine consultations,” Computers in Biology and Medicine, vol. 172, p. 108290, Apr. 2024, doi: 10.1016/j.compbiomed.2024.108290. [12] 蘇志民, “協助醫師問診的系統及方法,” TW201905934A, Feb. 01, 2019 Accessed: Jul. 03, 2024. [Online]. Available: https://patents.google.com/patent/TW201905934A/zh [13] K. Krishna, S. Khosla, J. Bigham, and Z. C. Lipton, “Generating SOAP Notes from Doctor-Patient Conversations Using Modular Summarization Techniques,” in Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), C. Zong, F. Xia, W. Li, and R. Navigli, Eds., Online: Association for Computational Linguistics, Aug. 2021, pp. 4958–4972. doi: 10.18653/v1/2021.acl-long.384. [14] A. Asai, S. Min, Z. Zhong, and D. Chen, “Retrieval-based Language Models and Applications,” in Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 6: Tutorial Abstracts), Y.-N. (Vivian) Chen, M. Margot, and S. Reddy, Eds., Toronto, Canada: Association for Computational Linguistics, Jul. 2023, pp. 41–46. doi: 10.18653/v1/2023.acl-tutorials.6. [15] K. Roy et al., “Using Large Pretrained Language Models for Answering User Queries from Product Specifications,” in Proceedings of the 3rd Workshop on e-Commerce and NLP, S. Malmasi, S. Kallumadi, N. Ueffing, O. Rokhlenko, E. Agichtein, and I. Guy, Eds., Seattle, WA, USA: Association for Computational Linguistics, Jul. 2020, pp. 35–39. doi: 10.18653/v1/2020.ecnlp-1.5. [16] L. Yang et al., “MixPAVE: Mix-Prompt Tuning for Few-shot Product Attribute Value Extraction,” in Findings of the Association for Computational Linguistics: ACL 2023, A. Rogers, J. Boyd-Graber, and N. Okazaki, Eds., Toronto, Canada: Association for Computational Linguistics, Jul. 2023, pp. 9978–9991. doi: 10.18653/v1/2023.findings-acl.633. [17] N. Ousidhoum, Z. Yuan, and A. Vlachos, “Varifocal Question Generation for Fact-checking,” in Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, Y. Goldberg, Z. Kozareva, and Y. Zhang, Eds., Abu Dhabi, United Arab Emirates: Association for Computational Linguistics, Feb. 2022, pp. 2532–2544. doi: 10.18653/v1/2022.emnlp-main.163. [18] A. Rani et al., “FACTIFY-5WQA: 5W Aspect-based Fact Verification through Question Answering,” in Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), A. Rogers, J. Boyd-Graber, and N. Okazaki, Eds., Toronto, Canada: Association for Computational Linguistics, Jul. 2023, pp. 10421–10440. doi: 10.18653/v1/2023.acl-long.581. [19] B. P. Majumder, S. Rao, M. Galley, and J. McAuley, “Ask what’s missing and what’s useful: Improving Clarification Question Generation using Global Knowledge,” in Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, K. Toutanova, A. Rumshisky, L. Zettlemoyer, D. Hakkani-Tur, I. Beltagy, S. Bethard, R. Cotterell, T. Chakraborty, and Y. Zhou, Eds., Online: Association for Computational Linguistics, Jun. 2021, pp. 4300–4312. doi: 10.18653/v1/2021.naacl-main.340. [20] Z. Zhang and K. Q. Zhu, “Diverse and Specific Clarification Question Generation with Keywords,” in Proceedings of the Web Conference 2021, Apr. 2021, pp. 3501–3511. doi: 10.1145/3442381.3449876. [21] Z. Wang, Y. Tu, C. Rosset, N. Craswell, M. Wu, and Q. Ai, “Zero-shot Clarifying Question Generation for Conversational Search,” in Proceedings of the ACM Web Conference 2023, Austin TX USA: ACM, Apr. 2023, pp. 3288–3298. doi: 10.1145/3543507.3583420. [22] X. Jiang, S. Wang, W. Liu, and Y. Yang, “Prediction of traditional Chinese medicine prescriptions based on multi-label resampling,” Journal of Electronic Business & Digital Economics, vol. 2, no. 2, pp. 213–227, Jan. 2023, doi: 10.1108/JEBDE-04-2023-0009. [23] W. Xu and Q. Wu, “A Novel TCM Prescription Recommendation Method based on Attention Factorization Machines,” Journal of Computing and Electronic Information Management, vol. 10, pp. 55–61, May 2023, doi: 10.54097/jceim.v10i3.8682. [24] X. Dong et al., “TCMPR: TCM Prescription recommendation based on subnetwork term mapping and deep learning,” in 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Feb. 2021, pp. 3776–3783. doi: 10.1109/BIBM52615.2021.9669588. [25] Z. Zhao et al., “PreGenerator: TCM Prescription Recommendation Model Based on Retrieval and Generation Method,” IEEE Access, vol. 11, pp. 103679–103692, 2023, doi: 10.1109/ACCESS.2023.3316219. [26] J. Qi, X. Wang, and T. Yang, “Traditional Chinese Medicine Prescription Recommendation Model Based on Large Language Models and Graph Neural Networks,” in 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), Feb. 2023, pp. 4623–4627. doi: 10.1109/BIBM58861.2023.10385489. [27] Y. Cui, Z. Yang, and X. Yao, “Efficient and Effective Text Encoding for Chinese LLaMA and Alpaca,” Feb. 22, 2024, arXiv: arXiv:2304.08177. doi: 10.48550/arXiv.2304.08177. [28] “A dataset for Chinese-English Terminology of Chinese Medicine.” Accessed: Oct. 13, 2024. [Online]. Available: https://www.scidb.cn/en/detail?dataSetId=798559406290632704 [29] Sturgeon D., “醫方集解 - 中國哲學書電子化計劃.” Accessed: Oct. 13, 2024. [Online]. Available: https://ctext.org/wiki.pl?if=gb&res=244407
指導教授	蔡宗翰林筱玫(Tzong-Han Tsai Hsiao-Mei Lin)	審核日期	2025-1-17
推文	facebook plurk twitter funp google live udn HD myshare reddit netvibes friend youpush delicious baidu
網路書籤	Google bookmarks del.icio.us hemidemi myshare

博碩士論文 111525009 詳細資訊