在機器維修人員的養成能力當中，最重要也最核心的能力就是問題的判斷與解決方法的推理能力，這些能力的養成並非一蹴可及，需要時間與經驗的累積。且同樣的問題發生，會因為不同的人有不同的經驗而有不同的故障排除與元件檢查模式，但並不是所有的處理模式都能夠真正找到根本原因(root cause)的所在，反而可能因為不知道機器原件的連接關係演變成一而再再而三的錯誤嘗試(Try and Error)，導致時間的浪費與機器產量的減少。因此，如何在最快的時間內利用之前維修過的紀錄與經驗或是利用根本的元件連接關係將機器異常之處找出來予以故障排除，就變成是相當重要且必須的事情。
本論文係使用IC封裝廠的機台維修紀錄資料庫為系統主要維修資料庫，配合上本體論概念，建構出元件與元件間之互相連接關係，並將兩者結合之後推論出最有可能之損壞或動作異常之元件，將之給予建議，讓使用者在遇到機器異常狀況時，可以知道該從何處開始維修與確認工作。若是有經驗較為不足或是未遇過發生異常的設備工程師遇到機器故障無法排除時，利用詢問的方式可以獲得之前所發生過類似的問題處理方式作為參考，進行元件的狀態檢查或是維修；若是該問題在之前從沒發生過或是未曾記錄在資料庫裡，系統將會依所建構的本體論進行推測，給予使用者一個或多個建議，建議機器維修人員可以從何處著手進行維修，減少機器維修人員錯誤嘗試的時間，更可以達到知識共享，增加設備維修工程師的經驗與能力，提高機台單位時間之產量。
問答系統(Question Answering System, QA system)是利用使用者所輸入的問句加以剖析，依據問句的語意給予最適合的答案。問答系統不但可以減少使用者尋找資料的時間，也可以達到將知識再利用的目的。

For equipment engineers’ training, the most important is the problem solving and reasoning ability. Such ability cannot be obtained with just exemplary case trainings in a very short time, but has to accumulate from many experiences in a very long time practice. Even an engineer with a rich experience may use different trouble shooting methods on different problems. But not all methods can find out a cause easily and precisely. Sometimes one has to repeat try-and-error due to unclear relation between components and wastes time and reduces productions. One possible way to get around this is to use maintenance records from others' experiences and the inherent relations of equip components.
This thesis proposes a method and implements a system that uses both the maintenance record databases of an IC assembly factory and the relations of equip components inferred from an ontology for suggesting the most possible problem solving procedures to equipment maintenance engineers. When a junior engineer that does not have much experience has an abnormal machine to repair, he can interact with this question- answering system to get suggestions for maintaining the machine. The maintenance records will accumulate when time goes by, and if a problem that does not occur before, this system will build an ontology relationship among components or states and make suggestions according to that. Using this platform can not only save time for equipment maintenance but also share knowledge in experience among engineers.

[1] Anne Kao and Stephen R. Poteet, “Natural Language Processing and Text Mining”, Springer, 2006.
[2] AnswerBus, http://www.answerbus.com/
[3] Ask.com, http://www.ask.com/
[4] ASQA, http://asqa.iis.sinica.edu.tw/
[5] Cheng-Wei Lee, Cheng-Wei Shih, Min-Yuh Day, Tzong-Han Tsai, Tian-Jian Jiang, Chia-Wei Wu, Cheng-Lung Sung, Yu-Ren Chen, Shih-Hung Wu, Wen-Lian Hsu, “Perspectives on Chinese Question Answering Systems”, in Proceedings of the Workshop on the Sciences of the Artificial (WSA 2005), Hualien, Taiwan, 2005.
[6] Chinese Knowledge Information Processing Group, “CKIP AutoTag”,Academic sinica, 1998.
[7] CKIP中文詞知識庫小組, http://ckip.iis.sinica.edu.tw/CKIP.
[8] Dieter Fensel, “Ontologies:A Silver Bullet for Knowledge Management and Electronic Commerce 2nd Edition”, Springer, 2003.
[9] Grigoris Antoniou and Frank van Harmelen, “A Semantiv Web Primer”, The MIT Press, 2004.
[10] Han, J. and Kamber M.,” Data mining: Concepts and techniques”,Morgan Kaufmann, 2001.
[11] Keh-Jiann Chen and Ming-Hong Bai, “UnknownWord Detection for Chinese by a Corpus-based Learning Method”, Computational Linguistics and Chinese Language Processing, vol. 3, no. 1, pp. 27-44, 1998.
[12] Keh-jiann Chen and Wei-Yun Ma, “A bottom-up Merging Algorithm for Chinese Unknow Word Extraction”, in Proceedings of ACL workshop on Chinese Language Processing, pp. 31-38, 2003.
[13] Mehrnoush Shamsfard and Ahmad Abdollahzadeh Barforoush , “Learning ontologies from natural language texts”, Internal Journal of Human-Computer Studies vol.60 pp.17-63, ELSEVIER, 2004.
[14] R. Neches, R. Fikes, T. Finin, T. Gruber, R. Patil, T. Senator, and W.R. Swartout, “ENABLING TECHNOLOGY FOR KNOELEDGE SHARING”, AI Magazine, vol. 12, pp. 36-56, 1991.
[15] Raj Sharman, Rajiv Kishore and Ram Ramesh, “Ontologies: A handbook of principles, concepts and applications in information systems”, Springer, 2007.
[16] Sheng-Yuan Yang, “Developing of an ontological interface agent with template-based linguistic processing technique for FAQ services”, Expert Systems with Applications, Vol.36, pp.4049–4060, ELSEVIER, 2009.
[17] STAR, http://start.csail.mit.edu/
[18] Stuart Russell and Peter Norvig, ”Artificial Intelligence: A Modern Approach 2/E”, Pearson, 2003.
[19] Sung-Shun Weng , Hsine-Jen Tsai, Shang-Chia Liu and Cheng-Hsin Hsu, “Ontology construction for information classification”, Expert Systems with Applications vol.31 pp.1–12, ELSEVIER, 2006.
[20] Sung-Shun Weng and Hui-Ling Chang, “Using ontology network analysis for research document recommendation, Expert Systems with Applications”, vol.34, pp.1857–1869, ELSEVIER, 2008.
[21] Sung-Shun Weng, Hsine-Jen Tsai, Shang-Chia Liu and Cheng-Hsin Hsu, “Ontology construction for information classification”, Elsevier, Expert Systems with Applications 31 (2006) 1–12.
[22] Wesley T.Chuang, Jihoon Yang, “Extracting Sentence Segment for Text Summarization﹕A Machine Learning Approach”, In ACM SIGIR 2000, pp.152-159, 2000.
[23] Wikipedia, http://en.wikipedia.org.
[24] Ying-Hong Wang, Cheng-Horng Liao, Shih-Hao Huang and Wen-Nan Wang, “Application Architecture of Semantic QA System”, Journal of computers, Vol. 17, No.3, 2006.
[25] 翁頌舜、洪湘貽, “以本體論為基之情境感知音樂推薦”, ICIM2009第二十屆國際資訊管理學術研討會, pp. 2130-2139, 2009。
[26] 許正欣, “語意網上自動化建構本體論之研究”, 天主教輔仁大學資訊管理學系碩士論文, 2004。
[27] 陳慶龍, “可提供語意搜尋之部分知識建構”, 國立成功大學工程科學系碩士論文, 2007。