在資訊大量擴充與爆炸的今日，加上資訊種類的繁多與複雜，所以更是難以找尋正確與所需的資料。而利用資訊檢索(Information Retrieval)與資訊擷取(Information Extraction)的方法，我們便可以易於在大量的資料中檢索與擷取重要的資訊。
問題答覆答系統結合了資訊檢索與資訊擷取，在大量的文件中找尋問題相關的內文，進而擷取其答案。資訊尋找方式通常是利用資訊檢索的技術，但資訊檢索所得的資訊過於廣泛且雜訊過多，所以加上資訊擷取的方法，可以把資訊精簡。但單純的加入資訊擷取與資訊檢索，真正感興趣的部分還是無法得知，這時就需要專有名詞(Name Entity)辨識我們感興趣的部分，並加以擷取。一般的資訊檢索與資訊擷取無法直接套用在問題回答系統，原因是問題與答案的種類繁多，而且涉及自然語言的格式與方法，加上隨字彙語義、語法不同，語句的表示法也會不同，所以大部分問題答覆系統都需要進一步的問題分類(Question Classification)與段落擷取(Passage Retrieval)技巧，並加上人所觀察出的經驗法則(Heuristic)來解決問題與答案間的關連性。而人的因素牽涉越多，所花的成本也隨之增大。也由於人類相關的知識介入，所牽涉的領域很廣，很難用一個通則涵蓋所有範圍。
而本篇所要設計的問題回答系統，即是利用已知的資訊加上分類演算法來建立系統模組，模組會自動學習如何找尋問題的答案。此種機器學習(Machine Learning)的技巧能讓系統面對未來可利用的訓練資料時，更能學習到重要資訊，而不需複雜的人為介入造成時間、人力成本的增加。這種以分類為基礎的問題回答系統是第一次被嘗試，而實驗也證明了其獨特性與優越性。

It is a world of information explosion nowadays. Due to the variety and the complexity of information, the accurate data becomes more difficult to search. Meanwhile, people may have tended to neglect some important information which appears shortly. By using Information Retrieval (IR) and Information Extraction (IE) techniques, it is beneficial for helping people to fetch accurate and important information within a large amount of databases more effectively.
A Question Answering System (QA system) combines both IR and IE techniques. It is able to search answers in documents of questions. Information Retrieval usually uses Document Retrieval to find the relevant documents, but the documents may have too much information and many noise. Hence, most QA Systems use question classification and passage retrieval to improve the system accuracy. Then, they use Name Entity to tag the proper noun they interested. Because QA systems involve linguistics studying, most of them use the observations of human efforts to create the relations between questions and answers. But more human efforts involve, more time and money spend.
This research of the QA System is designed to utilize the information that is already known. It includes classified questions and correct answer sentences. By adding Machine Learning techniques, our QA system integrates the information and classification-based methods. We can answer the question automatically without human efforts. It is the first time that QA systems use classification-based system architecture. And from our experiments, they prove that our QA system has its uniqueness and superiority.

[1] E. Voorhees, “The TREC-8 Question Answering Track Report,” Proceedings of the Eighth Text Retrieval Conference, 77, 1999.
[2] E. Voorhees. “Overview of the TREC-9 Question Answering Track,” Proceedings of the Ninth Text Retrieval Conference, 71, 2000.
[3] E. Voorhees, “Overview of the TREC 2001 Question Answering Track,” Proceedings of the Tenth Text Retrieval Conference, 42, 2001.
[4] E. Voorhees, “Overview of the TREC 2002 Question Answering Track,” Proceedings of the Eleventh Text Retrieval Conference, 2002.
[5] E. Voorhees, “Overview of the TREC 2003 Question Answering Track,” Proceedings of the Twelfth Text Retrieval Conference, 2003.
[6] C. J. Lin and H. H. Chen, “Description of Preliminary Results to TREC-8 QA Task,” Proceedings of the Eighth Text Retrieval Conference, 1999.
[7] C. J. Lin and W. C. Lin, “Description of NTU QA and CLIR System in TREC-9,” Proceedings of the Ninth Text Retrieval Conference, 2000.
[8] C. J. Lin and H. H. Chen, “Description of NTU System at TREC-10 QA Track,” Proceedings of the Tenth Text Retrieval Conference, 406, 2001.
[9] R.F.E. Sutcliffe, “Question Answering Using the DLT System at TREC 2002,” Proceedings of the Eleventh Text Retrieval Conference, 2002.
[10] M.A. Greenwood, I. Roberts, and R. Gaizauskas, “The University of Sheffield TREC 2002 Q&A System,” Proceedings of the Eleventh Text Retrieval Conference, 2002.
[11] M. Wu, X. Zheng, M. Duan, T. Liu, and T. Strzalkowski, “Questioning Answering By Pattern Matching, Web-Proofing, Semantic Form Proofing,” Proceedings of the Twelfth Text Retrieval Conference, 2003.
[12] A. Ittycheriah, M. Franz, and S. Roukos, “IBM''s Statistical Question Answering System--TREC-10,” Proceedings of the Tenth Text Retrieval Conference, 258, 2000.
[13] J. Prager, J. Chu-Carroll, K. Czuba, C. Welty, A. Ittycheriah, R. Mahindru, “IBM''s PIQUANT in TREC2003,” Proceedings of the Twelfth Text Retrieval Conference, 283, 2003.
[14] H. Yang and T.-S. Chua, “The Integration of Lexical Knowledge and External Resources for Question Answering,” Proceedings of the Eleventh Text Retrieval Conference, 2002.
[15] D. Moldovan, S. Harabagiu, R. Girju, P. Morarescu, F. Lacatusu, A. Novischi, A. Badulescu, and O. Bolohan, “LCC Tools for Question Answering,” Proceedings of the Tenth Text Retrieval Conference, 2001
[16] X. Li and D. Roth, “Learning Question Classifiers,” Proceedings of the 19th International Conference on Computational Linguistics, 2002.
[17] G. Salton and C. Buckley, “Improving retrieval performance by relevance feedback,” Journal of the American Society for Information Science, 41(4):288-297, 1990.
[18] T. Kudo, Y. Matsumoto, “Use of Support Vector Learning for Chunk Identification,” Proceedings of CoNLL, 2000
[19] G. G. Lee, J. Seo, S. Lee, H. Jung, B.-H. Cho, C. Lee, B.-K. Kwak, J. Cha, D. Kim, J. An, H. Kim, and K. Kim, “SiteQ: Engineering high performance QA system using lexico-semantic pattern matching and shallow NLP,” Proceedings of the Tenth Text Retrieval Conference, 2001.
[20] S. Tellex, B. Katz, J. Lin, A. Fernandes, and G. Marton, “Quantitative evaluation of passage retrieval algorithms for question answering,” Proceedings of the 26th annual international ACM SIGIR conference on Research and development in information retrieval, 41–47, 2003.
[21] M. Pasca and S. Harabagiu, “High-Performance Question Answering,” 24th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 366-374, 2001.
[22] K. Hacioglu and W. Ward, “Question Classification with Support Vector Machines and Error Correcting Codes,” Proceedings of HLT-NAACL, 2003.
[23] D. Zhang and W. S. Lee, “Question Classification using Support Vector Machines,” Proceedings of the 26th annual international ACM SIGIR conference on Research and development in information retrieval, 26-32, 2003.