臺灣博碩士論文加值系統

本研究的目的為在建立一個對話式機器人(Conversational Bot)輔助民眾於火災的疏散撤離。
民眾在火災發生時是否能順利疏散避難，取決於其行為與建築物特性和發展中的火災之間的相互作用。而當災害發生時，因為一般民眾對於火災的認知有限，在災害發生時難以輕易有效判斷現場狀況，面臨災害威脅者是否能有效於有限之時間與空間內進行避難逃生，具有不可言喻之重要性。本研究透過回顧以行人流相關理論進行之時間模擬相關文獻，依據火災求生專家的火場經驗談為基礎，再以對話式系統與決策樹推薦方法與人對談互動建立避難決策模式，並且探討該人機互動模式的適用性與限制。
在人機互動對話模式的限制條件上，本研究以火災疏散的時間當作避難生存為主要限制條件，運用自然語言處理(Natural Language Processing)及對話式人機介面(Conversational User Interfaces)，以FaceBook Messenger 為平台探討火災時對話式機器人對於一般人火災避難有效性之效益與行動決策之影響，期望能藉此研究提供避難逃生指引與防災演練參考。

In this study, an attempt was made to develop a conversational bot for assisting people evacuate in case of fire.
Whether civilians could adequately evacuate in the event of a fire depends on the dynamics between their behavior and the characteristics of buildings and burning fires.
When a fire occurs, people have limited knowledge about fire, which makes it difficult to assess the scenario in time while the fire is spreading. However, it is of paramount importance whether people can effectively escape within a limited time and space.
In this study, an evacuation decision-making model is proposed, based on literature review on evacuation simulations based on pedestrian flow correlation theory the experience of fire survival experts. Moreover, the interaction between the evacuation decision-making model, dialogue-based system, decision tree recommendation method, and the applicability and limitation of the human-computer interaction mode are discussed.
Under the limitation mode of human-computer interaction, this study used the time of evacuation as the main limitation of successful escape. Using Natural Language Processing and Conversational User Interfaces, Messenger was used as a platform to explore the impact of dialogue-type robots on fire evacuee patterns in the event of fire. It is hoped that the results of this study will be useful for establishing or reviewing guidelines or standard procedures for fire evacuation and fire drills.

目錄
摘要 i
Abstract ii
誌謝 iii
目錄 iv
圖目錄 vi
表目錄 vii
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機與目的 3
1.3 研究流程與架構 4
第二章 文獻回顧 5
2.1 對話式機器人(Chatbot)的定義及應用發展現狀 5
2.1.1自然語言處理與與語意理解 5
2.1.2 對話管理與架構 6
2.1.3 對話式機器人類別與應用 8
2.2 火災應變過程與受災者行為特性與決策分析 10
2.2.1 火災之危害特性與災例分析 10
2.2.2火災應變避難思維模式 11
2.2.3人員行動特性之避難與逃生決策意涵 12
2.3 科技接受模型（Technology Acceptance Model ，TAM） 15
2.4通訊軟體平台差異 17
第三章 系統開發與設計 19
3.1 系統架構與設計 19
3.2 對話式機器人之建構 20
3.2.1 開發環境與流程 21
3.2.2自然語言處理模組 21
3.2.3 使用者介面之互動設計 23
3.3 對話流程模組實作 25
3.4 火災情境假設及使用情境設計 29
3.5 Facebook Messenger Platform Design Kit 31
3.5.1 SDK &Webhook 31
3.6實驗設計及架構 39
3.6.1實驗設計 39
第四章 系統呈現與實驗設計 41
4.1系統呈現與操作流程說明 41
4.2 實驗目的 46
4.3實驗任務和程序 49
4.4分析探討 54
第五章 結論與建議 63
5.1 結論 63
5.2 研究限制與未來研究建議 64
參考文獻 65

參考文獻
1. End-to-End Task-Completion Neural Dialogue Systems
Xiujun Li, Yun-Nung Chen, Lihong Li, Jianfeng Gao ,Mar 2017
2. Quantification of behaviour for engineering design standards and escape time calculations EDA Purser, M Bensilum - Safety science, 2001
3. Xiujun Li, Zachary C Lipton, Bhuwan Dhingra, Lihong
Li, Jianfeng Gao, and Yun-Nung Chen. 2016.
A user simulator for task-completion dialogues. arXiv
preprint arXiv:1612.05688
4. Sungjin Lee. 2014. Extrinsic evaluation of dialog state
tracking and predictive metrics for dialog policy optimization.
In 15th Annual Meeting of the Special Interest Group on Discourse and Dialogue .
5. Lifeng Shang, Zhengdong Lu, Hang Li.2015.
Neural Responding Machine for Short-Text Conversation
6. Jia, J. (2009). CSIEC: A computer assisted English learning Chatbot based on textual knowledge and reasoning. Knowledge-Based Systems, 22(4), 249-255.
7. Mikolov, T., Karafiat, M., Burget, L., ´ Cernock ˇ y, J., and Khudan- ´ pur, S. (2010). Recurrent neural network based language model. In Proceedings of INTERSPEECH, vol. 2010, no. 9. International Speech Communication Association, 2010, pp. 1045–1048.
8. Victor W Zue and James R Glass. 2000. Conversational interfaces: Advances and challenges. Proceedings of the IEEE 88(8):1166–1180.
9. JZ Hernandez, JM Serrano - Expert Systems with Applications, 2001 - Elsevier
Knowledge-based models for emergency management systems
10. JG Doheny, JL Fraser - Expert Systems With Applications, 1996 - Elsevier
MOBEDIC—A decision modelling tool for emergency situations

11. Thomas, N. T. (2016, September). An e-business Chatbot using AIML and LSA. In Advances in Computing, Communications and Informatics (ICACCI), 2016 International Conference on (pp. 2740-2742). IEEE.
12. Charles Vaught, Ph.D., Michael J. Brnich, Jr., Launa G. Mallett, Ph.D., Henry P. Cole, Ed.D., William J. Wiehagen, Ronald S. Conti, Kathleen M. Kowalski, Ph.D., and Charles D. Litton
Behavioral and Organizational Dimensions of Underground Mine Fires
13. Winograd, T. (1972). Understanding natural language. Cognitive psychology, 3(1), 1-191.ts of language—syntax, semantics, and inference.
14. Weizenbaum, J. (1966). ELIZA—a computer program for the study of natural language communication between man and machine. Communications of the ACM, 9(1), 36-45.
15. Lokman, A. S., & Zain, J. M. (2010). One-match and all-match categories for keywords matching in chatbot. American Journal of Applied Sciences, 7(10), 1406-1411.
16. J Weizenbaum -MIT. Communications of the ACM, 1966
17. Dušek, Ondrej, and Filip Jurcıcek. "A Context-aware Natural Language Generator for Dialogue Systems." 17th Annual Meeting of the Special Interest Group on Discourse and Dialogue. 2016.
18. KAPLAN, R. (2016). Beyond the GUI: It’s Time for a Conversational User Interface.
19. On the Impacts of Utilizing Smartphones on Organizing Rescue Teams and Evacuation Procedures ;Parvaneh Sarshar ;2013
20. KAPLAN, R. (2016). Beyond the GUI: It’s Time for a Conversational User Interface. Retrieved from https://www.wired.com/2013/03/conversational-user-interface/
21. A stress model for people facing a fire ;GuylèneProulx
Journal of Environmental Psychology
Volume 13, Issue 2, June 1993, Pages 137-147

22. NFPA 72-National Fire Alarm and Signaling Code By National Fire Protection Association (NFPA)
中文書籍文獻：
23. 林金宏，活著離開（第二版） 2014。出版社：雙葉書廊
24. 國內民眾防火避難認知調查之研究-內政部建築研究所-吳秉宸、蕭嘉俊http://www.abri.gov.tw/tw/research/show/1201/p/print
網路資源：
25. StatCounter (gs.statcounter.com/)
26. BI Intelligence(www.businessinsider.com/research)
27. Chatbotsmagazine (https://chatbotsmagazine.com/)
28. Facebook messenger (https://messenger.fb.com/)
29. Api.ai (https://dialogflow.com/)