跳到主要內容

臺灣博碩士論文加值系統

(44.220.44.148) 您好!臺灣時間:2024/06/21 14:40
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:黃典良
研究生(外文):HUANG, DIAN-LIANG
論文名稱:不同NLP模型對意圖辨識聊天機器人效能影響之研究
論文名稱(外文):Research on the Performance of Different NLP Models for the Chatbot with Intent Recognition
指導教授:林嬿雯林嬿雯引用關係
指導教授(外文):Lin, Yen-Wen
口試委員:林嬿雯顧維祺蔡坤霖
口試委員(外文):Lin, Yen-WenKu, Wei-ChiCAI, KUN-LIN
口試日期:2020-07-27
學位類別:碩士
校院名稱:國立臺中教育大學
系所名稱:資訊工程學系
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:中文
論文頁數:57
中文關鍵詞:聊天機器人自然語言處理意圖辨識實體分類雙語評估學習
外文關鍵詞:ChatbotNLPIntent RecognitionEntity ClassificationBLEU
相關次數:
  • 被引用被引用:0
  • 點閱點閱:421
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:1
隨著資訊科技的快速進步,帶動了聊天機器人的發展。聊天機器人應用變得愈來愈流行,相關的應用,包括:服務型聊天機器人、陪伴型機器人與語音助理等。聊天機器人需要準確快速地回應。讓聊天機器人做出有效的回應是本文的研究目的。本文的研究重點放在利用NLP (Natural Language Processing)模型來改善聊天機器人的效能,使用者只要輸入特定的關鍵字就能與聊天機器人進行對話。
為了研究NLP對聊天機器人可用性的影響,本文進行一系列的實驗。本文實驗使用不同NLP模型來探討各個模型的效能表現。本研究試圖找到一種適合用於意圖辨識與實體分類的NLP模型,以提高聊天機器人的效能,包括:BLEU (BiLingual Evaluation Understudy)與成功率。實驗結果證明,使用NLP模型可以提高聊天機器人的可用性。

The promotion of information technology inspires the development of chatbots. Related applications of chatbots, including service chatbot, companion robots, and voice assistants; become acceptably. The chatbot has to respond accurately and quickly. It is the objective of this research to make the chatbot respond effectively. The research focuses on the usage of NLP (Natural Language Processing) model for improving the performance of the chatbot. The users can interact with the chatbot by entering the keywords.
To study the effects of NLP models on the usability of chatbots, a series of experiments are carried out in this thesis. Various NLP models are used in the experiments to investigate the performance of different models. The research try to find out a suitable NLP model for intent recognition and entity classification to improve the performance (namely BLEU (BiLingual Evaluation Understudy) and success rate) of the chatbot. The experimental results display that the usability of chatbots can be improved by using NLP models.

摘要 I
Abstract II
表目錄 V
圖目錄 VI
第一章 緒論 1
1.1 背景與動機 1
1.2 研究目標 1
1.3 論文架構 2
第二章 相關研究 3
2.1 聊天機器人 (Chatbot) 3
2.2 人工智慧 (Artificial Intelligence) 4
2.3 自然語言處理 (Natural Language Processing) 4
2.4 NLP模型 (NLP Model) 5
2.5 NLP.js 5
2.6 意圖辨識 (Intent Recognition) 6
2.7 實體分類 (Entity Classification) 7
2.8 問答系統 (Question and Answering System) 7
2.9 雙語評估學習 (BiLingual Evaluation Understudy) 8
2.10成功率 (Success Rate) 8
第三章 研究方法 9
3.1 問題定義 9
3.2 系統架構 10
3.3 實驗環境 10
3.4 研究方法 11
3.4.1 訓練模型的流程 11
3.4.2 創建資料集 12
3.4.3 模型的配置 13
3.4.4 配置模型參數並訓練 14
第四章 實驗與結果討論 17
4.1 NLP模型與意圖辨識 17
4.1.1 意圖辨識的實驗目的 17
4.1.2 意圖辨識的實驗設計 17
4.1.3 意圖辨識的實驗結果與討論 18
4.2 NLP模型與實體分類 30
4.2.1 實體分類的實驗目的 30
4.2.2 實體分類的實驗設計 31
4.2.3 實體分類的實驗結果與討論 31
4.3 聊天機器人滿意度測量 35
4.3.1聊天機器人的實驗目的 35
4.3.2聊天機器人的實驗設計 35
4.3.3聊天機器人的實測結果 36
4.4 小結 51
第五章 結論 52
5.1 結論 52
5.2 未來工作 52
參考文獻 53

[1]Shaikh, A., More, D., Puttoo, R., Shrivastav, S., & Shinde, S. "A survey paper on chatbots." IRJET 6.4 (2019).
[2]Rahman, A. M., Abdullah Al Mamun, and Alma Islam. "Programming challenges of chatbot: Current and future prospective." 2017 IEEE Region 10 Humanitarian Technology Conference (R10-HTC). IEEE, (2017).
[3]Human–Robot Interaction, available at: https://www.teksystems.com/en/insights/video/how-to-design-natural-language-intent-recognition-for-chatbots access (2020.7.10)
[4]Shawar, Bayan Abu, and Eric Atwell. "A corpus-based approach to generalising a chatbot system." Procesamiento del lenguaje natural 31 (2003).
[5]機器人如何理解人類語言, available at: https://research.sinica.edu.tw/nlp-natural-language-processing-chinese-knowledge-information/ access (2020.7.18)
[6]Setyawan, Muhammad Yusril Helmi, Rolly Maulana Awangga, and Safif Rafi Efendi. "Comparison of multinomial Naive Bayes algorithm and logistic regression for intent classification in chatbot." 2018 International Conference on Applied Engineering (ICAE). IEEE, (2018).
[7]Handoyo, E., Arfan, M., Soetrisno, Y. A. A., Somantri, M., Sofwan, A., & Sinuraya, E. W. "Ticketing Chatbot Service using Serverless NLP Technology." 2018 5th International Conference on Information Technology, Computer, and Electrical Engineering (ICITACEE). IEEE, (2018).
[8]Bhagwat, Vyas Ajay. "Deep Learning for Chatbots." (2018).
[9]Okuda, Takuma, and Sanae Shoda. "AI-based chatbot service for financial industry." Fujitsu Scientific and Technical Journal 54.2: 4-8 (2018).
[10]Ghorpade-Aher, J., Kontamwar, R., Kukreja, S., Karpe, T., & Kakkad, S. "An overview of NLP based Chatbot." (2019).
[11]王昊奮、邵浩, "自然語言理解基本技術, "中文自然語言處理實戰:聊天機器人與深度學習整合應用,1th ed.博碩文化, Country: TW,Abbrev. of Publisher, 2019.07 pp.2-9.
[12]自然語言處理, available at: https://lionbridge.ai/articles/how-to-build-a-social-ai-an-interview-with-nlp-researcher-thomas-wolf/ access (2020.7.10)
[13]Google BERT, available at: https://www.ithome.com.tw/news/133855 access (2020.7.10)
[14]OpenAI GPT-2, available at: http://baijiahao.baidu.com/s?id=1652093322137148754 access (2020.7.10)
[15]Transformer, available at: https://www.ithome.com.tw/news/138141 access (2020.7.10)
[16]NLP.js, available at: https://github.com/axa-group/NLP.js access (2020.6.30)
[17]基於意圖(Intent-based)與基於流程(flow-based)的Chatbot available at: https://medium.com/botbonnie/nlp-%E9%97%9C%E6%96%BC%E6%84%8F%E5%9C%96%E5%BC%8F-intent-based-%E8%B7%9F%E6%B5%81%E7%A8%8B%E5%BC%8F-flow-based-chatbot-1555fbfc322c access (2020.7.10)
[18]Nigam, Amber, Prashik Sahare, and Kushagra Pandya. "Intent Detection and Slots Prompt in a Closed-Domain Chatbot." 2019 IEEE 13th International Conference on Semantic Computing (ICSC). IEEE, (2019).
[19]Sanchez, Sebastian. "An Intent-Aware Chatbot for Cybersecurity Recommendation. ", (2020).
[20]Indulkar, Ajinkya Pradeep, Srivatsan Varadharajan, and Krishnamurthy Nayak. "Neural Intent Recognition for Question-Answering System." International journal of innovative research in technology 5.3: 54-65 (2018).
[21]Kim, Jae-Ho, In-Ho Kang, and Key-Sun Choi. "Unsupervised named entity classification models and their ensembles." COLING 2002: The 19th International Conference on Computational Linguistics (2002).
[22]實體分類,available at: https://tpu.thinkpower.com.tw/tpu/articleDetails/1800?gclid=EAIaIQobChMIqvj5lYi_6gIVRXZgCh1-pQn5EAMYASAAEgKbpfD_BwE access (2020.7.10)
[23]Sarkar, S., Rao, V., Baala Mithra, S. M., & Rao, S. V. "NLP Algorithm Based Question and Answering System." Seventh International Conference on Computational Intelligence, Modeling and Simulation (2015).
[24]Dutta, Sourav, and Dietrich Klakow. "Evaluating a neural multi-turn chatbot using BLEU score." (2019).
[25]Ilievski, V., Musat, C., Hossmann, A., & Baeriswyl, M."Goal-oriented chatbot dialog management bootstrapping with transfer learning." arXiv preprint arXiv:1802.00500 (2018).
[26]Google BERT Model, available at: https://github.com/Nagakiran1/Extending-Google-BERT-as-Question-and-Answering-model-and-Chatbot/blob/master/modeling.py access (2020.6.30)
[27]OpenAI GPT-2 Model, available at: https://github.com/DoktorHolmes/Maxwell/blob/master/gpt2bot/model.py access (2020.6.30)
[28]Transformer Model, available at: https://github.com/minhhahao/botNet/blob/master/bot/model.py access (2020.6.30)
[29]Dimovski, M., Musat, C., Ilievski, V., Hossmann, A., & Baeriswyl, M. "Submodularity-inspired data selection for goal-oriented chatbot training based on sentence embeddings." arXiv preprint arXiv:1802.00757 (2018).
[30]Variani, E., Bagby, T., McDermott, E., & Bacchiani, M."End-to-end training of acoustic models for large vocabulary continuous speech recognition with tensorflow." (2017).
[31]Kelley, R., Tavakkoli, A., King, C., Ambardekar, A., Nicolescu, M., & Nicolescu, M. "Context-based bayesian intent recognition." IEEE Transactions on Autonomous Mental Development 4.3: 215-225 (2012).
[32]Kim, Jae-Ho, In-Ho Kang, and Key-Sun Choi. "Unsupervised named entity classification models and their ensembles." COLING 2002: The 19th International Conference on Computational Linguistics (2002).
[33]Dropout, available at: https://medium.com/@a5560648/dropout-5fb2105dbf7c access (2020.8.10)
[34]Batch Size, available at: https://codertw.com/%E7%A8%8B%E5%BC%8F%E8%AA%9E%E8%A8%80/557816/ access (2020.8.10)
[35]Epochs, available at: https://kknews.cc/zh-tw/code/kban458.html access (2020.8.10)
[36]train.py, available at: https://github.com/xuemzhan/Intent-recognition/blob/master/cnn/train.py access (2020.7.22)
[37]BLEU Score, available at: https://www.itread01.com/content/1545325022.html access (2020.6.30)
[38]test.py, available at: https://github.com/Sentdex/chatbotrnd/blob/master/chat_question_examples.pyaccess (2020.6.30)
[39]Chatbot-1.py, available at: https://github.com/henry16lin/Voice-Assistance/blob/master/voice_assistant.py access (2020.6.30)
[40]NLP.js, available at: https://github.com/axa-group/NLP.js access (2020.6.30)

電子全文 電子全文(網際網路公開日期:20250817)
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊