Skip navigation

DSpace

機構典藏 DSpace 系統致力於保存各式數位資料(如:文字、圖片、PDF)並使其易於取用。

點此認識 DSpace
DSpace logo
English
中文
  • 瀏覽論文
    • 校院系所
    • 出版年
    • 作者
    • 標題
    • 關鍵字
    • 指導教授
  • 搜尋 TDR
  • 授權 Q&A
    • 我的頁面
    • 接受 E-mail 通知
    • 編輯個人資料
  1. NTU Theses and Dissertations Repository
  2. 電機資訊學院
  3. 資訊工程學系
請用此 Handle URI 來引用此文件: http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/100182
完整後設資料紀錄
DC 欄位值語言
dc.contributor.advisor陳縕儂zh_TW
dc.contributor.advisorYun-Nung Chenen
dc.contributor.author林宗聖zh_TW
dc.contributor.authorTzung-Sheng Linen
dc.date.accessioned2025-09-24T16:46:15Z-
dc.date.available2025-09-25-
dc.date.copyright2025-09-24-
dc.date.issued2025-
dc.date.submitted2025-08-14-
dc.identifier.citation[1] G. Anderson, E. Hart, D. Gkatzia, and I. Beaver. An open intent discovery evaluation framework. In Proceedings of the 25th Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 760–769, 2024.
[2] I. Casanueva, T. Temčinas, D. Gerz, M. Henderson, and I. Vulić. Efficient intent detection with dual sentence encoders. In Proceedings of the 2nd Workshop on Natural Language Processing for Conversational AI, pages 38–45, 2020.
[3] J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 4171–4186, 2019.
[4] T. Gao, X. Yao, and D. Chen. SimCSE: Simple contrastive learning of sentence embeddings. arXiv preprint arXiv:2104.08821, 2021.
[5] M. Hong, Y. Song, D. Jiang, W. Ng, Y. Sun, and C. J. Zhang. Dial-in LLM: Human-aligned dialogue intent clustering with LLM-in-the-loop. arXiv preprint arXiv:2412.09049, 2024.
[6] S. Larson, A. Mahendran, J. J. Peper, C. Clarke, A. Lee, P. Hill, J. K. Kummerfeld, K. Leach, M. A. Laurenzano, L. Tang, et al. An evaluation dataset for intent classification and out-of-scope prediction. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 1311–1316, 2019.
[7] H. Lin, L. Ma, J. Zhu, L. Xiang, Y. Zhou, J. Zhang, and C. Zong. CSDS: A fine-grained Chinese dataset for customer service dialogue summarization. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 4436–4451, 2021.
[8] P. Liu, Y. Ning, K. K. Wu, K. Li, and H. Meng. Open intent discovery through unsupervised semantic clustering and dependency parsing. arXiv preprint arXiv:2104.12114, 2021.
[9] J. MacQueen. Some methods for classification and analysis of multivariate observations. In Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1: Statistics, volume 5, pages 281–298. University of California Press, 1967.
[10] A. C. Müller and S. Guido. Introduction to Machine Learning with Python: A Guide for Data Scientists. O’Reilly Media, Inc., 2016.
[11] J. Pennington, R. Socher, and C. D. Manning. GloVe: Global vectors for word representation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1532–1543, 2014.
[12] N. Reimers and I. Gurevych. Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 3982–3992, 2019.
[13] X. Shen, Y. Sun, Y. Zhang, and M. Najmabadi. Semi-supervised intent discovery with contrastive learning. In Proceedings of the 3rd Workshop on Natural Language Processing for Conversational AI, pages 120–129, 2021.
[14] H. Xu, B. Liu, L. Shu, and P. Yu. Open-world learning and application to product classification. In The World Wide Web Conference, pages 3413–3419, 2019.
[15] J. Xu, P. Wang, G. Tian, B. Xu, J. Zhao, F. Wang, and H. Hao. Short text clustering via convolutional neural networks. In Proceedings of the 1st Workshop on Vector Space Modeling for Natural Language Processing, pages 62–69, 2015.
[16] H. Zhang, H. Xu, T.-E. Lin, and R. Lyu. Discovering new intents with deep aligned clustering. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 14365–14373, 2021.
[17] H. Zhang, H. Xu, X. Wang, F. Long, and K. Gao. A clustering framework for unsupervised and semi-supervised new intent discovery. IEEE Transactions on Knowledge and Data Engineering, 36(11):5468–5481, 2023.
[18] S. Zhang, J. Yang, J. Bai, C. Yan, T. Li, Z. Yan, and Z. Li. New intent discovery with attracting and dispersing prototype. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 12193–12206, 2024.
[19] Y. Zhang, H. Zhang, L.-M. Zhan, X.-M. Wu, and A. Lam. New intent discovery with pre-training and contrastive learning. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 256–269, 2022.
-
dc.identifier.urihttp://tdr.lib.ntu.edu.tw/jspui/handle/123456789/100182-
dc.description.abstract在開放世界的對話系統中,使用者輸入的語句可能包含訓練階段未曾出現的意圖,因此「新意圖探勘」成為一項關鍵任務。
本研究提出一套創新的新意圖探勘與命名架構,結合對比學習與凍結的大型語言模型嵌入向量,在保留預訓練模型泛化能力的同時,學得具辨識性的語意表示。我們設計的模組化流程將嵌入生成、語意表示學習與聚類三個階段明確分離,具備高度彈性,可因應未來大型語言模型嵌入向量或聚類技術的演進而彈性替換或擴充。為提升實務應用中的可解釋性,我們進一步引入自動命名機制,為新探勘出的意圖指派具可讀性的語意標籤。此一面向在過去文獻中常被忽略,或受限於固定命名模板,然而對於實際部署至關重要。
此外,我們亦提出一套基於命名的評估方法,量化模型所產生之意圖名稱與人工標註名稱的一致性,直接反映意圖探勘結果的語意品質。我們在多項標準資料集及一套具挑戰性的真實多輪對話語音轉文字之資料上進行實驗,驗證本方法在效能、泛化能力與穩健性上的優勢。
zh_TW
dc.description.abstractOpen-world dialogue systems must address utterances with evolving intents not seen during training, making new intent discovery (NID) essential.
This paper proposes a novel NID framework that combines contrastive learning with frozen large language model (LLM) embeddings to generate discriminative representations without compromising pretrained generalization. Our modular pipeline explicitly separates embedding generation, representation learning, and clustering, enabling flexible integration of future advances in LLMs and clustering techniques. To improve practical interpretability, we introduce an automated naming mechanism that assigns human-readable labels to newly discovered intents, which is often an overlooked but critical component in real-world deployment.
We further present a naming-based evaluation framework to directly assess the quality of intent discovery by measuring alignment between system-generated and human-annotated intent names. Experiments on benchmarks and a challenging in-house dataset
of multi-turn ASR transcripts demonstrate the effectiveness, generalizability, and robustness of our approach.
en
dc.description.provenanceSubmitted by admin ntu (admin@lib.ntu.edu.tw) on 2025-09-24T16:46:15Z
No. of bitstreams: 0
en
dc.description.provenanceMade available in DSpace on 2025-09-24T16:46:15Z (GMT). No. of bitstreams: 0en
dc.description.tableofcontentsAcknowledgements i
摘要 ii
Abstract iii
Contents v
List of Figures vii
List of Tables viii
Chapter 1 Introduction 1
1.1 Introduction 1
Chapter 2 TENN (Tracking Evolving Needs and Naming) 5
2.1 TENN (Tracking Evolving Needs and Naming) 5
2.1.1 Problem Definition 5
2.1.2 Proposed Framework 6
2.1.3 Training – Contrastive Learning for Utterance Projection 7
2.1.4 Testing – Projection and Clustering 8
2.1.5 Testing – Seen Intent Assignment 9
2.1.6 Testing – Unseen Intent Discovery and Naming 9
Chapter 3 Experiments 11
3.1 Experiments 11
3.1.1 Datasets 11
3.1.1.1 BANKING77 11
3.1.1.2 CSDS 12
3.1.1.3 IN-HOUSE BANKCALL 12
3.1.2 Experimental Setup 13
3.1.3 Baselines 13
3.1.4 Metrics 16
3.1.5 New Intent Discovery Results 17
3.1.6 New Intent Discovery and Naming Results 19
Chapter 4 Conclusion 22
4.1 Conclusion 22
References 23
-
dc.language.isoen-
dc.subject意圖命名zh_TW
dc.subject對比學習zh_TW
dc.subject新意圖探勘zh_TW
dc.subject對話系統zh_TW
dc.subject大型語言模型zh_TW
dc.subjectDialogue Systemsen
dc.subjectIntent Namingen
dc.subjectLarge Language Modelsen
dc.subjectContrastive Learningen
dc.subjectNew Intent Discoveryen
dc.title基於人與人對話進行未知使用者意圖探勘、建構及命名方法zh_TW
dc.titleUnseen User Intent Discovery and Induction from Human-Human Conversationsen
dc.typeThesis-
dc.date.schoolyear113-2-
dc.description.degree碩士-
dc.contributor.oralexamcommittee孫紹華;陳重吉zh_TW
dc.contributor.oralexamcommitteeShao-Hua Sun;Chung-Chi Chenen
dc.subject.keyword新意圖探勘,對比學習,大型語言模型,意圖命名,對話系統,zh_TW
dc.subject.keywordNew Intent Discovery,Contrastive Learning,Large Language Models,Intent Naming,Dialogue Systems,en
dc.relation.page26-
dc.identifier.doi10.6342/NTU202504217-
dc.rights.note同意授權(限校園內公開)-
dc.date.accepted2025-08-15-
dc.contributor.author-college電機資訊學院-
dc.contributor.author-dept資訊工程學系-
dc.date.embargo-lift2028-12-31-
顯示於系所單位:資訊工程學系

文件中的檔案:
檔案 大小格式 
ntu-113-2.pdf
  未授權公開取用
3.28 MBAdobe PDF檢視/開啟
顯示文件簡單紀錄


系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。

社群連結
聯絡資訊
10617臺北市大安區羅斯福路四段1號
No.1 Sec.4, Roosevelt Rd., Taipei, Taiwan, R.O.C. 106
Tel: (02)33662353
Email: ntuetds@ntu.edu.tw
意見箱
相關連結
館藏目錄
國內圖書館整合查詢 MetaCat
臺大學術典藏 NTU Scholars
臺大圖書館數位典藏館
本站聲明
© NTU Library All Rights Reserved