簡易檢索 / 詳目顯示

研究生: 傅秉文
Fu, Ping-Wen
論文名稱: 具真人輔導學習機制之客製化銀髮族對話系統
Customized Dialogue System for Senior People with Human Assisted Learning Mechanism
指導教授: 王駿發
Wang, Jhing-Fa
學位類別: 碩士
Master
系所名稱: 電機資訊學院 - 電機工程學系
Department of Electrical Engineering
論文出版年: 2015
畢業學年度: 103
語文別: 英文
論文頁數: 64
中文關鍵詞: 對話系統句子相似度計算Wizard of Oz機器學習客製化照護系統
外文關鍵詞: Dialogue System, Sentence Similarity, Wizard of Oz, Machine Learning, Customized, Care System
相關次數: 點閱:107下載:0
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報
  • 本論文提出一個具有真人輔導學習機制的客製化對話系統架構,將其應用於銀髮族照護系統。該系統可根據不同使用者可以做出口語對話、影音媒體播放、資訊檢索、協助聯絡等行為回饋。在對話系統原型設計採用有限狀態基礎模型實現,並藉由提出的PRCBRB句子相似度演算法實現快速的對話系統架構。於對話策略階段根據Wizard of Oz理論加入真人輔助學習機制,此機制可將系統無法處理的對話狀況交給支援者處理,系統紀錄這個過程,並對於對話系統劇本修改。在客製化機制部分,本論文自訂了個人資訊知識表示格式及兩種知識檢索方法,可讓回饋動作根據使用者的登入資訊做出調整,達到客製化功能。最後於介面設計上則根據銀髮族對於科技的使用習慣做出使用者介面調整,實務層面達到相映的成果。實驗部分,對於PRCBRB演算法做出分析,以MCDC語料庫做為測試資料,在inside-test下,正確率可達到97%,運算效能則相較於其他類似演算法快5倍。在Outside-test的部分,實際對於31位使用者(其中包含10位65歲以上的銀髮族)使用狀況評估12種使用者需求任務的任務達成率,一般使用者可達到86.7%,銀髮族為64%,在MOS平均主觀評測分數方面,一般使用者達到4.4,銀髮族可達到為4.3,在實務上是適用於銀髮族的。

    In this thesis, we proposed a customized dialogue system with human assisted learning mechanism is proposed to realize the orange technologies of warming-care system for seniors. The system has the capability of spoken dialogue, audio and video media playing, information retrieval and remote assistance connection etc. Moreover, the system feedback strategy is adapted according to the enrolled user information for customized request. The dialogue system prototype is designed based on Finite-state-script model and the proposed POS-Reference-Column-Based-Row-Based (PRCBRB) sentence similarity algorithm. Human assisted learning mechanism is added to stage of dialogue strategy according to Wizard-of-Oz design methodology. If the system fails to recognize how to respond the user request, the system will automatically call for assistance from supporter. Then feedback of supporter is relayed to user by system. Assistance of supporter is converted into the new script and added to script database by learning mechanism. The aforementioned customized mechanism is fundamentally realized by the following proposed techniques, including, the personal information knowledge representation format and two kinds of knowledge retrieval methods. In the experiments, the Mandarin Conversational Dialogue Corpus (MCDC) corpus is applied for testing and analyzing the efficiency of the PRCBRB algorithm. In the inside-test, the accuracy can achieve to 97.3% accuracy and the computation cost is 5 times faster than the other algorithms. In the outside-test, the task completion rate (TCR) of 12 kinds of tasks for 31 users including 10 seniors over 65 are evaluated. For the general user and the senior users, TCR can be reached to 86.7% and 64%, respectively. In Mean Opinion Score evaluation, the average MOS score is up to 4.4 for the general use, and up to 4.3 for the senior user. In light of above experimental results, the proposed system is proven the usability for the seniors in the real life.

    中文摘要 I Abstract II 誌謝 IV Contents V Table List VII Figure List IX Chapter 1 Introduction 1 1.1 Background 1 1.2 Motivation 3 1.3 Objectives 4 1.4 Contribution 4 1.5 Organization 5 Chapter 2 Related Works 6 2.1 Overview of Dialogue System 6 2.1.1 Finite State-based 6 2.1.2 Frame-based 7 2.1.3 Mixed-initiative 8 2.2 Sentence Similarity Measures 8 2.2.1 Sentence Similarity based on Word Set 9 2.2.2 Sentence Similarity based on Word Vector 9 2.2.3 Sentence Similarity based on Dynamic Time Warping 10 2.2.4 Whole-matching-plane-based (WMPB) 10 2.3 Learning Method for Dialogue System 11 2.3.1 Markov Decision Process Method 11 2.3.2 Q-Learning Method 12 2.3.3 WOZ 13 2.4 Knowledge Representation research 14 2.4.1 Semantic Network 14 2.4.2 Object-Attribute-Value Triples (OAV) 15 2.5 Overview of relationship between Senior People and Technology 15 Chapter 3 Integrated Customized dialogue system with Human Assisted Learning Mechanism 17 3.1 Architecture Overview 17 3.2 Dialogue system prototype design 18 3.2.1 Pre-processing 18 3.2.2 Finite-state script format 20 3.2.3 PRCBRB for dialog management 22 3.2.4 Dialogue Strategy 31 3.3 Human Assisted Learning Mechanism (HALM) 32 3.3.1 Setting of Human Assisted 33 3.3.2 Learning mechanism activated 35 3.3.3 Script paradigm restructuring 37 3.4 Customized Mechanism 38 3.4.1 Biographical knowledge Format 39 3.4.2 Knowledge Extraction 40 3.4.3 Task processor 43 3.4.4 Feedback content adjustment 45 3.5 System construction 45 3.5.1 User-side interface 46 3.5.2 Supporter-side interface 47 Chapter 4 Experiments and Results 48 4.1 Sentence Similarity Performance Evaluation 48 4.1.1 Corpus 48 4.1.2 Pre-processing 49 4.1.3 Result 51 4.2 Objective Performance Evaluation 51 4.2.1 Life Care Feedback Task 51 4.2.2 Experimental environment and flow 52 4.2.3 Result 54 4.3 Subjective Performance Evaluation 54 4.3.1 Evaluation points setting 54 4.3.2 Result 56 Chapter 5 Conclusion and Future Work 57 5.1 Conclusions 57 5.2 Future works 57 References 58 Appendix-A 62

    [1] "The State of the Art of Dialogue Systems", From SAM – Socialising Around Media, [Online]. Available: http://samproject.net/the-state-of-the-art-of-dialogue-systems/
    [2] "Cortana vs Siri vs Google Now vs Blackberry Assistant - fight!", From AAWP – All About Windows Phone, [Online]. Available: http://allaboutwindowsphone.com/features/item/20697_Cortana_vs_Siri_vs_Google_Now-.php
    [3] "Turing test", From Wikipedia, [Online]. Available: https://en.wikipedia.org/wiki/Turing_test
    [4] "Computer passes Turing Test for first time by convincing judges it is a 13-year-old boy", From The Telegraph, [Online]. Available: http://www.telegraph.co.uk/technology/news/10884839/Computer-passes-Turing-Test-for-the-first-time-after-convincing-users-it-is-human.html
    [5] "Cognitive science", From Wikipedia, [Online]. Available: https://en.wikipedia.org/wiki/Cognitive_science
    [6] 黃財丁,"由統計資料看人口老化問題", From Research Portal, [Online]. Available: http://portal.stpi.narl.org.tw/index/article/37
    [7] "United Nations Population Fund, UNFPA", From UNFPA - United Nations Population Fund, [Online]. Available: http://www.unfpa.org/
    [8] U. Cort´es, R. Annicchiaricoe, "Assistive technologies for the disabled and for the new generation of senior citizens: The e-tools architecture," AI Commun., vol. 16, pp. 193–207, 2003.
    [9] Yunxin Zhao, "Speech-Recognition Technology in Health Care and Special-Needs Assistance [Life ciences]", IEEE Signal Processing Magazine, Volume:26, Issue: 3,pp. 87 – 90, 2009
    [10] Yu-De Chu, and Chia-Hui Chang. "MAGEN: An Adaptive Conversational System based on Terms", Association for Computational Linguistics and Chinese Language Processing (ACLCLP), Taiwan, 2006
    [11] Maozu Guo, Yang Liu, and Jacek Malec "A new Q-learning Algorithm Based on the Metropolis Criterion" IEEE Transaction On Systems Man And Cybernetics, Volume:34, Issue: 5, pp. 2140 - 2143, 2004
    [12] Esther Levin, Roberto Pieraccini and Wieland Eckert, "A Stochastic Model of Human-Machine Interaction for Learning Dialogue Strategy", IEEE Transactions On Speech And Audio Processing, Volume. 8, No. 1, 2000
    [13] Yi-Hsuan Lin, Shan-Ju Lin Chang, "Digital Divides Revisited: A Process View of the Acquisitions of Information and Communication Technology (ICT) Skills by the Elderly", 圖書資訊學研究 3:pp. 75-102, June 2009
    [14] J. F. Allen, D. K. Byron, M. Dzikovska, G. Ferguson, L. Galescu, and A. Stent, "Towards Conversational Human-Computer Interaction," AI Magazine, 2001.
    [15] M. F. McTear, "Spoken Dialog Technology: Enableing the Conversational User Interface," ACM Computing Surveys, vol. 34, pp. 90-169, Mar. 2002.
    [16] C. J. Lee, E. F. Huang, and J. K. Chen, "A Multi-keyword Spotter for the Application of the TL Phone Directory Assistant Service" Workshop on Distributed System Technologies & Applications, pp. 197-202, 1997.
    [17] Y. C. Lin, T. H. Chiang, H. M. Wang, C. Peng, and C. Chang, "The Design of Mandarin Chinese Spoken Dialogue System," International Conference on Spoken Language, vol. 1, pp. 230–233, 1998.
    [18] H. Aust, M. Oerder, F. Seide, and V. Steinbiss, "The Philips automatic train timetable information system," Speech Communication, vol. 17, pp. 249-262, 1995.
    [19] M. J. Chen, "Intention Extraction for Intelligent Medical Query System," National Cheng Kung University, Master dissertation, Jun. 2003.
    [20] Patidar A. K., Agrawal, J. and Mishra, N., "Analysis of Different Similarity Measure Functions and their Impacts on Shared Nearest Neighbor Clustering Approach", International Journal of Computer Applications, Vol.40, No.16, pp.1-5, February 2012
    [21] Junsheng Zhang, Yunchuan Sun, Huilin Wang, Yanqing He , "Calculating Statistical Similarity between Sentences", Journal of Convergence Information Technology, Volume 6, Number 2. February 2011
    [22] Natt-och-Dag, J., Regnell, B., Carlshamre, P., Andersson, M., and Karlsson, J., "Evaluating Automated Support for Requirements Similarity Analysis in Market-Driven Development", 7th Int. Workshop on Requirements Engineering: Foundation for Software Quality, , Interlaken, Switzerland, June 4-5 2001
    [23] Huang, A., "Similarity measures for text document clustering", Proceedings of the Sixth New Zealand Computer Science Research Student Conference (NZCSRSC2008), Christchurch, New Zealand, pp. 49–56. , 2008
    [24] G. Salton and M. McGill. Introduction to modern information retrieval. McGraw-Hill New York, 1983.
    [25] D. Gildea and D. Jurafsky. "Automatic labeling of semantic roles. Computational Linguistics", 28(3):, pp.245–288, 2002.
    [26] "Dynamic time warping", From Wikipedia, [Online]. Available: https://en.wikipedia.org/wiki/Dynamic_time_warping
    [27] Xiaoying Liu, Yiming Zhou, Ruoshi Zheng , "Sentence Similarity based on Dynamic Time Warpin", International Journal of Semantic Computing - IJSC , pp. 250-256, 2007
    [28] P.-C. Lin, J.-C. Wang, J.-F. Wang, and L.-C. Wen, "Design and portable device implementation of feature-based partial matching algorithms for personal spoken sentence retrieval", IET Signal Process, vol. 1, no. 3, pp. 139-149, September 2007.
    [29] J.-W. Wu, "Friendly elderly care spoken dialogue system based on Improved spontaneous keyword spotting and learnable jumping dialogue decision tree", masters dissertation, Univ. of Cheng Kung, Taiwan, 2012.
    [30] 蔡金翰,"語音對話系統和對話策略之研究", 國立交通大學,電信工程學系,碩士論文,2005,07
    [31] Yoshimasa Ohmoto, Hiroki Ohashi, Akihiro Takahashi, and Toyoaki Nishida, "Capture and Express Behavior Environment (CEBE) for Realizing Enculturating Human-Agent Interaction", Culture and Computing, Volume 6259, pp. 41-54 2010.
    [32] "ICIE: Immersive Collaborative Interaction Environment", [Online]. Available: http://www.ii.ist.i.kyoto-u.ac.jp/?page_id=1580
    [33] Newell, A. "The Knowledge Level", Artificial Intelligence. 18. pp. 87-127. Jan, 1982.
    [34] S.S. Tseng & G.J Hwang, "Knowledge Representation", [Online]. Available: http://140.118.59.33/gjhwang/KE&ES-chap2-%20%E7%9F%A5%E8%AD%98%E8%A1%A8%E7%A4%BA%E6%B3%952007-9-17A.pdf
    [35] Yang, Y.-K, "Behind the inheritance relations in a semantic network", Southeastcon '90. Proceedings., IEEE, pp.289 - 296 vol.1, 1-4 Apr 1990
    [36] Rich, E., "Artificial Intelligence", MeGraw-Hill, 1983
    [37] "Mycin", From Wikipedia, [Online]. Available: https://en.wikipedia.org/wiki/Mycin
    [38] Chung-Ming Huang; Chang, Y.-I.; Liu, M.T., "A computer-aided protocol design by production systems approach", Selected Areas in Communications, IEEE Journal on, Volume: 8, Issue: 9 ,Pages: 1748 - 1762, Year: 1990
    [39] 張書銓。公共圖書館實施銀髮族資訊利用教育之研究:以台北市立圖書館為例。未出版之碩士論文。國立台灣大學圖書館學研究所,臺北市。1998
    [40] Vicki L. Hanson., "Web access for elderly citizens", pp. 14 - 18, 2001
    [41] Czaja, S. J. & Lee, C. C., "The impact of the Internet on older adults", In Charness & Schaie (Ed.) Impact of Technology on Successful Aging, pp.113-133. NY: Springer Publishing., 2003.
    [42] 楊玉明,高齡者電腦態度之研究,未出版之碩士論文,國立高雄師範大學成人教育研究所,高雄市,1997
    [43] Selwyn, N., Gorard, S., Furlong, J., & Madden, L., "Older adults' use of information and communications technology in everyday life", Ageing & Society, 23, pp. 561-582, 2003.
    [44] Juznic, P., Blazic, M., Mercun, T., Plestenjak, B., & Majcenovic, D.,"Who says that old dogs cannot learn new tricks?", New Library World, 107(1226/1227), pp.332-345, 2006
    [45] Gerd Paul, Christian Stegbauer, "Is the digital divide between young and elderly people increasing?", Volume 10, Number 10 - 3 October 2005
    [46] 莊淑富,高齡者參與電腦學習動機之研究,未出版之碩士論文,國立中正大學成人及繼續教育研究所,嘉義縣,2002
    [47] Millward, P., "The grey digital divide: Perception, exclusion and barriers of access to the Internet for older people", Volume 8, Number 7 - 7, July 2003
    [48] "中文斷詞系統", From 中文斷詞系統, [Online]. Available: http://ckipsvr.iis.sinica.edu.tw/
    [49] "中研院平衡語料庫詞類標記集", [Online]. Available: http://mcdm.ntcu.edu.tw/profchwu/course/ml/%E6%95%99%E6%9D%90/%E6%96%87%E4%BB%B6%E5%88%86%E9%A1%9E/%E4%B8%AD%E7%A0%94%E9%99%A2%E8%A9%9E%E6%80%A7%E5%AE%9A%E7%BE%A9.pdf
    [50] Chomsky, Noam (1959). "On Certain Formal Properties of Grammars". Information and Control 2: pp.137–167.
    [51] "工研院文字轉語音Web服務", From Industrial Technology Research Institute,[Online]. Available: http://tts.itri.org.tw/development/web_service_api.php
    [52] "Maximum likelihood", From Wikipedia, [Online]. Available: https://en.wikipedia.org/wiki/Maximum_likelihood
    [53] "Imitation", From Parenting Counts, [Online]. Available: http://www.parentingcounts.org/professionals/parenting-handouts/information-for-parents-imitation.pdf
    [54] Alotaibi, M.B.; Rigas, D.I., "Fostering the User Interface Acceptance in Customer Relationship Management: A Multimedia-aided Approach", Information Technology: New Generations (ITNG), 2012 Ninth International Conference on, Page 796 – 801, 16-18 April 2012
    [55] S. C. Tseng, "Processing spoken Mandarin corpora", Traitement automatique des langues. Special Issue: Spoken Corpus Processing, vol. 45, no. 2, pp. 89-108, 2004.
    [56] Kyungduk Kim, Cheongjae Lee, Sangkeun Jung and Gary Geunbae Lee, "A Frame-Based Probabilistic Framework for Spoken Dialog Management Using Dialog Examples", Proceedings of the 9th SIGdial Workshop on Discourse and Dialogue, pages 120–127, Columbus, June 2008
    [57] "Method for subjective determination of transmission quality", ITU-T Recommendation. P.800, 1996.

    無法下載圖示 校內:2020-08-27公開
    校外:不公開
    電子論文尚未授權公開,紙本請查館藏目錄
    QR CODE