(3.237.97.64) 您好!臺灣時間:2021/03/03 01:49
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

我願授權國圖
: 
twitterline
研究生:周台亮
研究生(外文):CHOU, TAI-LIANG
論文名稱:基於長短期記憶遞迴模型和生成對抗式網路的任務型聊天機器人
論文名稱(外文):A Task-oriented Chatbot Based on LSTM and Generative Adversarial Networks
指導教授:薛幼苓 博士
指導教授(外文):HSUEH, YU-LING
口試委員:吳昇歐陽振森
口試委員(外文):WU, SUNOUYANG, CHEN-SEN
口試日期:2020-07-07
學位類別:碩士
校院名稱:國立中正大學
系所名稱:資訊工程研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2020
畢業學年度:108
語文別:英文
論文頁數:57
中文關鍵詞:深度學習聊天機器人服務機械人自然語言處理句子生成對話管理
外文關鍵詞:Deep LearningChatbotsService RobotsNatural Language ProcessingSentence GenerationDialogue Management
相關次數:
  • 被引用被引用:0
  • 點閱點閱:118
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
隨著深度學習技術的快速進步,聊天機器人已發生了革命性的改變,在通訊軟體裡的應用也隨之普及。 毫無疑問,聊天機器人是人類與機器互動的一種新方式。 然而,大多數典型的聊天機器人都是作為一個簡單的問題回答系統,用公式化的答案來回答問題。 傳統的聊天機器人通常採用基於檢索的模型,而這種模型需要大量的對話資料來檢索各種對話者的意圖。 因此,如何用較少的對話資料來生成一個有對話多樣性的聊天機器人模型是非常重要的。 本研究中,我們提出了基於生成對抗網路生成句子的方法來建構聊天機器人模型。 本模型之結構包含一個生成器,能依照不同的編碼生成不同的句子,包含行動碼、對話管理碼和人物碼。 在本模型中裡還有一個判別器,用於判斷生成的句子和原始數據的優劣。 在生成器中,我們結合了注意模型以追蹤句子狀態與使用雙向長短期記憶序列模型來提取句子的資訊。 在判別器中,我們計算了三種類型的回饋分數,分別為重覆句給出較低的獎勵分數和為具有多樣性的句子給出較高的獎勵分數。 通過實驗驗證了本模型的有效性,與現有的方法相比,本模型生成的句子更加多樣化,信息量也更加豐富。
Thanks to the advancements in deep learning, chatbots can be seen tremendously on messaging applications. Undoubtedly, a chatbot is a new way of interaction between humans and machines. However, most of the chatbots act as a simple question answering system that responses with formulated answers. Traditional conversational chatbots usually adopt a retrieved-based model which requires a large amount of conversational data for retrieving various intents. Hence, training a chatbot model that uses less conversational data to generate more diverse dialogue is very desirable. we propose a method to build a chatbot by a sentence generation model that generates sequence sentences based on the generative adversarial network. The architecture of our model contains a generator that generates a diverse sentence with a corresponding action code, a dialogue manage code, and a persona code, and a discriminator that judges the sentences between the generated and the raw data. In the generator, we combine the attention model that responses for tracking conversational states with the sequence-to-sequence model using hierarchical long-short term memory to extract sentence information. For the discriminator, we calculate three types of rewards to assign low rewards for repeated sentences and high rewards for diverse sentences. Extensive experiments are presented to demonstrate the utility of our model which generates more diverse and information-rich sentences than those of the existing approaches.
1 Introduction 1
2 Related Work 4
2.1 Rule-based and Retrieval-based Models . . . . . . . . . . . . . . . . . . 4
2.2 Finite State Machine . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.3 Generative Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3 Background Knowledge Overview 10
3.1 Word Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.1.1 LTP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1.2 ICTCLAS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1.3 jieba . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1.4 THULAC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.1.5 CKIP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2 Word Embedding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2.1 Continuous Bag-Of-Words and Skip-Gram . . . . . . . . . . . . 12
3.2.2 Embeddings From Language Models . . . . . . . . . . . . . . . 13
3.3 Recurrent Neural Networks . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.4 Long Short Term Memory . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.5 Reinforcement Learning . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.6 Sequence Generative Adversarial Networks . . . . . . . . . . . . . . . . 17
3.7 Attention Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.8 Diversity-Promoting GAN . . . . . . . . . . . . . . . . . . . . . . . . . 20
4 The Chatbot System Architecture 22
4.1 Dictionary Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . 23
4.2 Language Pre-processing . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.2.1 Word Segmentation . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.2.2 Word Embedding . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.3 Problem Formulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
4.4 Deep Learning Models . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.4.1 Generator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
4.4.2 Discriminator . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
4.5 Reward . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.5.1 Word-level Reward . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.5.2 Sentence-level Reward . . . . . . . . . . . . . . . . . . . . . . . 32
4.5.3 Information-level Reward . . . . . . . . . . . . . . . . . . . . . 32
4.6 Policy Gradient Training . . . . . . . . . . . . . . . . . . . . . . . . . . 33
5 Experiments 34
5.1 Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
5.2 Baseline Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.2.1 Sequence-to-sequence Learning (Seq2Seq): . . . . . . . . . . . . 38
5.2.2 Sequence Generative Adversarial Nets with Policy Gradient (SeqGAN): . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.2.3 Diversity Promoting GAN (DPGAN): . . . . . . . . . . . . . . . 38
5.3 Training Details . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
5.4 Evaluation Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
5.4.1 BLEU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5.4.2 GLEU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
5.4.3 METEOR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
5.4.4 ROUGE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
5.4.5 CIDEr . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
5.5 Experimental Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
6 Conclusion 50
7 Future Work 52

[1] “Siri.” https://www.apple.com/tw/siri/.
[2] “Build natural and rich conversational experiences.” https://dialogflow.
com/.
[3] “wit.ai.” https://wit.ai/.
[4] “Language understanding (luis).” https://www.luis.ai/.
[5] J. Li, W. Monroe, A. Ritter, D. Jurafsky, M. Galley, and J. Gao, “Deep reinforcement
learning for dialogue generation,” in Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, (Austin, Texas), pp. 1192–1202,
Association for Computational Linguistics, nov 2016.
[6] L. Yu, W. Zhang, J. Wang, and Y. Yu, “Seqgan: Sequence generative adversarial nets
with policy gradient,” in Thirty-First AAAI Conference on Artificial Intelligence,
2017.
[7] M. Schuster and K. K. Paliwal, “Bidirectional recurrent neural networks,” IEEE
transactions on Signal Processing, vol. 45, no. 11, pp. 2673–2681, 1997.
[8] W. J. Clancey, “Dialogue management for rulebased tutorials,” in Proceedings of the
6th International Joint Conference on Artificial Intelligence - Volume 1, IJCAI’79,
(San Francisco, CA, USA), p. 155–161, Morgan Kaufmann Publishers Inc., 1979.
[9] A. Bartl and G. Spanakis, “A retrieval-based dialogue system utilizing utterance and
context embeddings,” in Proceedings of 2017 16th IEEE International Conference
on Machine Learning and Applications (ICMLA), pp. 1120–1125, 2017.
[10] J. D. Williams, K. Asadi, and G. Zweig, “Hybrid code networks: practical and efficient end-to-end dialog control with supervised and reinforcement learning,” arXiv
preprint arXiv:1702.03274, 2017.
[11] X. Li, Y.-N. Chen, L. Li, J. Gao, and A. Celikyilmaz, “End-to-end task-completion
neural dialogue systems,” arXiv preprint arXiv:1703.01008, 2017.
[12] I. Sutskever, O. Vinyals, and Q. V. Le, “Sequence to sequence learning with neural
networks,” in Proceedings of advances in neural information processing systems,
pp. 3104–3112, 2014.
[13] J. Li, W. Monroe, T. Shi, S. Jean, A. Ritter, and D. Jurafsky, “Adversarial learning
for neural dialogue generation,” in Proceedings of the 2017 Conference on Empirical
Methods in Natural Language Processing, (Copenhagen, Denmark), pp. 2157–2169,
Association for Computational Linguistics, Sept. 2017.
[14] I. V. Serban, A. Sordoni, Y. Bengio, A. Courville, and J. Pineau, “Building endto-end dialogue systems using generative hierarchical neural network models,” in
Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, AAAI’16,
p. 3776–3783, AAAI Press, 2016.
[15] J. Xu, X. Ren, J. Lin, and X. Sun, “Diversity-promoting GAN: A cross-entropy
based generative adversarial network for diversified text generation,” in Proceedings
of the 2018 Conference on Empirical Methods in Natural Language Processing,
(Brussels, Belgium), pp. 3940–3949, Association for Computational Linguistics,
2018.
[16] 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),
(Minneapolis, Minnesota), pp. 4171–4186, Association for Computational Linguistics, June 2019.
[17] “Ckip.” https://ckip.iis.sinica.edu.tw/demo/.
[18] “Ckip.” https://ckip.iis.sinica.edu.tw/service/
ckiptagger/.
[19] “Ckip.” https://github.com/ckiplab/ckiptagger.
[20] W. Liu, Y. Zheng, S. Chawla, J. Yuan, and X. Xing, “Discovering spatio-temporal
causal interactions in traffic data streams,” in Proceedings of the 17th ACM SIGKDD
international conference on Knowledge discovery and data mining, pp. 1010–1018,
2011.
[21] M. E. Peters, M. Neumann, M. Iyyer, M. Gardner, C. Clark, K. Lee, and
L. Zettlemoyer, “Deep contextualized word representations,” arXiv preprint
arXiv:1802.05365, 2018.
[22] T. Mikolov, M. Karafiat, L. Burget, J. ´ Cernock ˇ y, and S. Khudanpur, “Recurrent neu- `
ral network based language model,” in Proceedings of eleventh annual conference
of the international speech communication association, 2010.
[23] S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Comput.,
vol. 9, pp. 1735–1780, Nov. 1997.
[24] R. S. Sutton and A. G. Barto, “Reinforcement learning: An introduction,” 2011.
[25] K. Yu, C. Dong, L. Lin, and C. Change Loy, “Crafting a toolchain for image restoration by deep reinforcement learning,” in Proceedings of the IEEE conference on
computer vision and pattern recognition, pp. 2443–2452, 2018.
[26] K. Zhou, Y. Qiao, and T. Xiang, “Deep reinforcement learning for unsupervised
video summarization with diversity-representativeness reward.,” in Proceedings of
the thirty-second AAAI conference on artificial intelligence (S. A. McIlraith and
K. Q. Weinberger, eds.), pp. 7582–7589, AAAI Press, 2018.
[27] C. Huang, S. Lucey, and D. Ramanan, “Learning policies for adaptive tracking with
deep feature cascades,” 2017 IEEE International Conference on Computer Vision
(ICCV), pp. 105–114, 2017.
[28] M. Arjovsky, S. Chintala, and L. Bottou, “Wasserstein gan,” arXiv preprint
arXiv:1701.07875, 2017.
[29] J.-Y. Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired image-to-image translation
using cycle-consistent adversarial networks,” in Proceedings of the IEEE international conference on computer vision, pp. 2223–2232, 2017.
[30] T. Luong, H. Pham, and C. D. Manning, “Effective approaches to attention-based
neural machine translation,” in Proceedings of the 2015 Conference on Empirical
Methods in Natural Language Processing, (Lisbon, Portugal), pp. 1412–1421, Association for Computational Linguistics, Sept. 2015.
[31] G. Klein, Y. Kim, Y. Deng, J. Senellart, and A. Rush, “OpenNMT: Open-source
toolkit for neural machine translation,” in Proceedings of ACL 2017, System Demonstrations, (Vancouver, Canada), pp. 67–72, Association for Computational Linguistics, July 2017.
[32] “ltp-cloud.” https://www.ltp-cloud.com.
[33] “heroku.” https://www.heroku.com/.
[34] “Emotibot.” http://www.emotibot.com/zh-tw/story.html?n=75.
[35] T. Emerson, “The second international Chinese word segmentation bakeoff,” in Proceedings of the Fourth SIGHAN Workshop on Chinese Language Processing, 2005.
[36] “Ictclas.” http://ictclas.nlpir.org/.
[37] “Jieba.” https://github.com/fxsjy/jieb/.
[38] “Thulac.” http://thulac.thunlp.org/.
[39] R. Y. Rubinstein and D. P. Kroese, The cross-entropy method: a unified approach to combinatorial optimization, Monte-Carlo simulation and machine learning. Springer Science & Business Media, 2013.
[40] R. J. Williams, “Simple statistical gradient-following algorithms for connectionist
reinforcement learning,” Machine learning, vol. 8, no. 3-4, pp. 229–256, 1992.
[41] Y. Wu, M. Schuster, Z. Chen, Q. V. Le, M. Norouzi, W. Macherey, M. Krikun,
Y. Cao, Q. Gao, K. Macherey, et al., “Google’s neural machine translation system: Bridging the gap between human and machine translation,” arXiv preprint
arXiv:1609.08144, 2016.
[42] K. Papineni, S. Roukos, T. Ward, and W.-J. Zhu, “Bleu: a method for automatic
evaluation of machine translation,” in Proceedings of the 40th Annual Meeting of
the Association for Computational Linguistics, (Philadelphia, Pennsylvania, USA),
pp. 311–318, Association for Computational Linguistics, July 2002.
[43] “dgk-lost-conv.” https://github.com/skdjfla/dgk_lost_conv.
[44] “Doubanconversaioncorpus.” https://archive.org/details/
DoubanConversaionCorpus.
[45] K. P. Murphy, Machine learning: a probabilistic perspective. 2012.
[46] J. C. Duchi, E. Hazan, and Y. Singer, “Adaptive subgradient methods for online learning and stochastic optimization,” Journal of Machine Learning Research,
vol. 12, pp. 2121–2159, 2010.
[47] A. Lavie and A. Agarwal, “Meteor: An automatic metric for mt evaluation with
high levels of correlation with human judgments,” in Proceedings of the Second
Workshop on Statistical Machine Translation, StatMT ’07, (USA), p. 228–231, Association for Computational Linguistics, 2007.
[48] C.-Y. Lin, “Rouge: A package for automatic evaluation of summaries,” p. 10, 01
2004.
[49] R. Vedantam, C. L. Zitnick, and D. Parikh, “Cider: Consensus-based image description evaluation,” 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4566–4575, 2015.
[50] S. Robertson, “Understanding inverse document frequency: On theoretical arguments for idf,” Journal of Documentation - J DOC, vol. 60, pp. 503–520, 10 2004.
[51] Q. Chen, Z. Zhuo, and W. Wang, “Bert for joint intent classification and slot filling,”
arXiv preprint arXiv:1902.10909, 2019.
[52] J. Vig and K. Ramea, “Comparison of transfer-learning approaches for response
selection in multi-turn conversations,” in Proceedings of workshop on DSTC7, 2019.
[53] Q. Chen, X. Zhu, Z.-H. Ling, S. Wei, H. Jiang, and D. Inkpen, “Enhanced LSTM
for natural language inference,” in Proceedings of the 55th Annual Meeting of the
Association for Computational Linguistics (Volume 1: Long Papers), (Vancouver,
Canada), pp. 1657–1668, Association for Computational Linguistics, July 2017.
[54] A. Radford, K. Narasimhan, T. Salimans, and I. Sutskever, “Improving language
understanding with unsupervised learning,” Technical report, OpenAI, 2018.
[55] T. M. Lai, Q. H. Tran, T. Bui, and D. Kihara, “A simple but effective bert
model for dialog state tracking on resource-limited systems,” arXiv preprint
arXiv:1910.12995, 2019.
電子全文 電子全文(網際網路公開日期:20230715)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔