本論文在建立具自然語言處理機制之數位學習虛擬助教系統，讓使用者能以口語化的方式詢問並獲得答案，而不需要經過特別的學習過程。我們提出層級式同義字詞庫的方法判斷問題分類，再以最長符合方式判斷中文斷詞，並改善其遇到相同長度字元時無法正確區分的問題。最後再由問題型態與查詢對應模組，以使用者問題與系統內部問答集比較相似度方式篩選出問題型態，並依問題型態回應正確的答案。
此外，一個完整的自然語言處理機制，除了前端的問題分析、查詢產生與結果反應外，後端的系統建置與維護也是極為重要。由於國軍人員異動頻繁，且大部分的系統管理人員都是由沒有資訊背景的人員兼任，以人工撰寫程式的方式管理維護系統往往造成許多問題。本論文提出以參數化建立系統的概念，利用異動圖產生查詢來建立問答系統相對應的問答集，讓系統管理維護更加容易。系統管理人員只需要進行相關屬性的參數設定，即可完成大部份的系統建置工作，避免系統因為管理人員的離職或退伍而難以延續，可以簡化設計、開發、維護及管理本系統。
本文章首先介紹如何以設定參數的方式建置問答系統相對應的問答集，接著說明本系統主要組成元件及系統雛型；最後我們提出實驗評估報告，證實本系統能達成簡化系統設定、減少錯誤率、加速系統建置且查詢正確率及系統查詢回應時間均在使用者可以接受的範圍之內。

This thesis is building up the system of e-learning and virtual teaching assistant with natural language processing mechanism, allow users can inquiry and acquire answers in oral way, and have no need any special learning process. This system judge the problems by classified experience rule and co-operate with the layers type of synonym saurus, to judge a break Chinese phrase in way of Longest Matching, and improve the problem of differentiation about the same length words. Finally, compare the problem type and enquire corresponding module with the user’s problem and the system’s internal Q&A likeness, and respond the correct answer according to problem type.
In addition, the problem analysis, enquiring generation and result responding of forehead in the integrated mechanism of natural language is very important as well as the system establishment and maintenance behind. Because of army personnel was posted away with the increasing frequency, and most of system managers is part-time and without information background, many problems in system maintenance was caused by artificial programming. This thesis builds up the concept of the system by parameter, create question and answer collection which opposite to system by using changing diagram, to make the maintenance of system more easily. The system manager can complete most work of the system establishments only by setting the parameters of related properties, avoiding system failed to last due to the manager’s retired, so that can simplify design, development, maintenance and management of system.
First of all, this article introduces how to establish the Q&A collection which opposite to the Q&A system by setting parameter. Then, explain the main component and model of system. The experiment estimation report was advanced finally, it approve that system can simplify system setting, reduce error rate, speed system establishment and correction of enquiry, let the range of responding time can be acceptted by use.

誌謝 ii
摘要 iii
ABSTRACT iv
目錄 v
表目錄 viii
圖目錄 ix
1. 前言 1
1.1 研究背景 1
1.2 研究動機 2
1.3 研究目的 4
1.4 論文架構 5
2. 文獻探討 6
2.1 自然語言問答系統與資訊檢索系統的差異 6
2.2 自然語言問答系統的分類 7
2.3 自然語言處理方法 7
2.3.1 中文斷詞 7
2.3.2 關鍵詞擷取 9
2.3.3 問題分類 9
2.4 Web自然語言問答系統 11
2.4.1 START(SynTactic Analysis Using Reversible Transformations) 11
2.4.2 NSIR 13
2.4.3 AnswerBus 14
2.5 相關研究分析與探討 16
2.5.1 An Analysis Of The AskMSR Question-Answering System 16
2.5.2 語意處理之中文自然語言擷取介面 16
2.5.3 概念式自動問答系統 17
2.5.4 中文自然語言查詢自動轉換為SQL之研究 18
2.5.5 Schema-Based Natural Language Semantic Mapping 19
2.5.6 下一代的數位學習環境－智慧型網路虛擬助教系統 20
2.5.7 物件導向網路學習平台之虛擬助教 21
2.5.8 國軍萬用資訊系統 22
2.6 綜合分析結果 22
3. 以參數化方式建置問答系統 25
3.1 異動圖 25
3.2 異動圖建置主要模組 26
3.3 異動圖限制條件 36
4. 系統架構 37
4.1 系統主要架構 37
4.2 中文斷詞模組 38
4.3 關鍵詞擷取模組 39
4.4 問題分類模組 41
4.5 問題型態與查詢對應模組 44
5.實驗與討論 46
5.1 實驗 46
5.1.2 實驗內容 47
5.1.3 實驗結果 48
5.2 討論 53
6.結論與未來研究方向 55
6.1 結論 55
6.2 未來研究方向 56
參考文獻 57
自傳 61

[1]http://www.iwillnow.org/。
[2]http://fms.cto.doh.gov.tw。
[3]http://linc.hinet.net/
[4]黃國禎，“下一代的數位學習環境－智慧型網路虛擬助教系統的研究”， 國立臺南大學數位學習科技系研究報告，台南，2005。
[5]蔡燿全、楊棠堯，“語意處理之中文自然語言擷取介面—網際網路平台” 商管科技季刊，第二卷，第一期，第45-65頁，2001。
[6]沈天佐、林川傑、陳信希，“以網際網路內容為基礎之問答系統Why 問句研究”，第十五屆計算語言學研討會論文集，新竹，第211-229頁，2003。
[7]楊燕珠、呂俊彥，“中文自然語言查詢自動轉換為SQL之研究”，2003數位生活與網際網路科技研討會，新竹，2003。
[8]陳意芬，“概念式自動問答系統”，碩士論文，國立交通大學資訊科學研究所，新竹，2003。
[9]蔣河山，“物件導向網路學習平台之虛擬助教”，碩士論文，國防大學理工學院資訊科學研究所，桃園，2005。
[10]歐陽玉慧，“國軍萬用資訊系統”，碩士論文，國防大學理工學院資訊科學研究所，桃園，2006。
[11]http://www.powerset.com/。
[12]http://www.parc.com/。
[13]Heinecke, J., and Toumani, F., “A Natural Language Mediation System for E-Commerce Applications: An Ontology-Based Approach,” Proceedings of Second International Semantic Web Conference, Sanibel Island, Florida, pp. 500-516, 2003.
[14]Lee, C. W., Shih, C. W., Day, M. Y., Tsai, T. H., Jiang, T. J., Wu, C. W., Sung, C. L., Chen, Y. R., Wu, S. H., and Hsu, W. L., “Perspectives on Chinese Question Answering Systems,” Proceedings of The Workshop on the Sciences of the Artificial, Hualien, pp. 113-128, 2005.
[15]http://www.languagecomputer.com/。
[16]http://www.answers.com/bb/。
[17]http://qa.iis.sinica.edu.tw/。
[18]黃居仁、陳克健、陳鳳儀、魏文真、張麗麗，“資訊處理用中文分詞規範設計理念及規範內容”，語言文字應用，第一期，第92-100頁，1997。
[19]曾元顯，“關鍵詞自動擷取技術與相關詞回饋”，中國圖書館學會會報第五十九期，第59-64頁，1997。
[20]Dan, M., Marius, P., Sanda, H., and Mihai, S., “Performance Issues and Error Analysis in An Open-Domain Question Answering System,” ACM Transactions on Information Systems (TOIS) , Vol. 21, Issue 2, pp. 133-154, 2003.
[21]Sutcliffe, R. F. E., Jia, X., and Mulcahy, M., “ Chinese Question Answering Using the DLT System at NTCIR 2005,” Proceedings of NTCIR-5 Workshop, Tokyo, 2005.
[22]Chen, J., Li, R., and Li, F.,“Chinese Information Retrieval Using Lemur: NTCIR-5 CIR Experiments at UNT,” Proceedings of NTCIR-5 Workshop, Tokyo, 2005.
[23]Kwok, K. L., Deng, P., Dinstl, N., and Choi, S., “NTCIR-5 English-Chinese Cross Language Question-Answering Experiments Using PIRCS,” Proceedings of NTCIR-5 Workshop, Tokyo, 2005.
[24]Ellen, R., and Michael, T., “A Rule-Based Question Answering System for Reading Comprehension Tests,” Proceedings of the ANLP/NAACL 2000 Workshop on Reading Comprehension Tests as Evaluation for Computer-Based Language Understanding Systems, Seattle, pp. 13-19, 2000.
[25]Lapedriza, À., and Jordi, V., “Open N-grams and Discriminant Features in Text World: An Empirical Study,” Proceedings of the Seventh Catalan Conference on Artificial Intelligence, Amsterdam, 2004.
[26]Cavnar, W. B. ,and Trenkle, J. M., “N-gram-Based Text Categorization. Proceedings of SDAIR-94,” Proceedings of the Third Annual Symposium on Document Analysis and Information Retrieval, Las Vegas, pp. 161-175, 1994.
[27]Peng, F., Schuurmans, D., and Wang, S., “Augmenting Naive Bayes Classifiers with Statistical Language Models,” Information Retrieval, Vol. 7 , Issue 3-4, pp. 317-345, 2004.
[28]Peng, F., and Schuurmans, D., “Combining Naive Bayes and N-Gram Language Models for Text Classification,” Proceedings of The Twenty-Fifth European Conference on Information Retrieval Research, Pisa, pp. 335-350, 2003.
[29]Meretakis, D., and Wüthrich, B., “Extending naïve Bayes Classifiers Using Long Itemsets,” Proceedings of the Fifth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Diego, pp. 165-174, 1999.
[30]Katz, B., Felshin, S., Yuret, D., Ibrahim, A., Lin, J., Marton, G., McFarland, A. J., and Temelkuran, B., “Omnibase: Uniform Access to Heterogeneous Data for Question Answering,” Proceedings of the seventh International Workshop on Applications of Natural Language to Information Systems, Stockholm, pp. 230-234, 2002.
[31]http://start.csail.mit.edu/。
[32]http://www.csail.mit.edu/。
[33]http://www.imdb.com/。
[34]http://www.cia.gov/redirects/ciaredirect.html。
[35]Radev, D. R., Fan, W., Qi, H., Wu, H., and Grewal A., “Probabilistic Question Answering on The Web,” Journal of the American Society for Information Science and Technology, Vol. 56, Issue 6, pp. 571-583, 2005.
[36]http://tangra.si.umich.edu/clair/NSIR/html/nsir.cgi。
[37]http://www.umich.edu/。
[38]Zheng, Z., “AnswerBus Question Answering System,” Proceeding of 2002 Human Language Technology Conference, San Diego, pp. 24-27, 2002.
[39]http://www.answerbus.com/index.shtml。
[40]Brill, E., Dumais, S., and Banko, M., “An Analysis of the AskMSR Question-Answering System,” Proceedings of The ACL-02 Conference on Empirical Methods in Natural Language Processing, Philadephia, pp. 257-264, 2002.
[41]Stratica, N., and Desai B. C., “Schema-Based Natural Language Semantic Mapping,” Proceedings of the Nineth International Conference on Applications of Natural Language to Information Systems (NLDB 2004), Salford, pp. 103-113, 2004.
[42]陳丙堃、陳昆鼎、張鳳蘭、王文洲，“物件導向網路學習平台”，兩岸機電暨產學合作學術研討會，新竹，2005。