對研究疾病的治療來說，電子醫囑中的時序資訊極為重要。時序資訊可分為日期、時間、頻率、期間四種主要類型。本研究分為兩主要部分。在第一部分中，我們實作了一套時序資訊識別系統。利用命名實體辨識技術，自電子醫囑的自然語言語句中，辨認出時序資訊，並加以分類並正規化。因應四類時序資訊，我們採用條件隨機域機器學習方法、Stanford CoreNLP工具及規則式方法並行運作，在國際知名的i2b2資料集上，可達到80%的F1-Measure。在第二部分中，我們採用科技接受模式與資訊系統成功模式兩種模型分析標示時序資訊的醫囑輸入系統，並邀請醫院護理師參與使用者調查。結果顯示，護理師普遍認同系統標示時序資訊的新功能。在管理意涵上，當純文字的醫囑標示出時序資訊，可帶來下列好處：(1) 護理部門管理階層可使用時序資訊來分析護理師工作內容，稽核護理師工作的品質。(2) 標示時序資訊可提高護理師對醫囑的理解度，確保病人受到正確地治療與照護。(3) 可用於醫囑執行完成率的計算。本研究成功地結合資訊系統實作與管理層面分析，對此議題之後續研究，將有重要的參考價值。

For treatment of diseases, temporal information in electronic prescriptions is extremely important. Temporal information can be divided into four main types: the date, time, frequency, period. This study has two main parts. In the first part, we construct a temporal named entity identification system that recognize temporal named entities from natural language sentences in electronic prescriptions and normalize these named entities. By hybridizing a CRF-based system, the Stanford CoreNLP system and our rule-based system, we achieve an F1-Measure of 80% on the well-known i2b2 dataset. In the second part, we use two models TAM and D&M to analyze the user acceptance of adding temporal information to physician order input systems. Over 50 hospital nurses were invited to participate in our user study. The results show that nurse practitioners generally agree that adding temporal information can improve their working efficiency. In management perspectives, highlighted temporal information brings the following benefits. (1) For care sector managers, temporal information can be used to analyze the working quality of nurses. (2) For nurses, highlighted temporal information can improve their understanding of doctor's orders, ensuring that patients receive proper treatment and care. (3) Identified temporal information can be used to calculate the completion rate of prescription orders. This study successfully combines the construction of an information system and the analysis in management perspectives, presenting the great reference value for follow-up studies in this topic.

第一章	緒論	1
1.1	研究背景	1
1.2	研究動機	2
1.3	研究問題與目的	3
1.4	研究範圍與限制	6
1.5	論文架構	7
第二章	文獻探討	9
2.1	實證醫學	9
2.2	ISO-8601	9
2.3	Regular Expressions	10
2.4	命名實體辨識（Named Entity Recognition）	12
2.5	i2b2 shared task in 2012	16
2.6	命名實體正規化（Named Entity Normalization）	18
2.7	科技接受度模型（Technology Acceptance Model）	21
2.8	資訊系統成功模式	22
2.9	本章小結	24
第三章	研究方法與系統架構	25
3.1	研究方法	25
3.2	系統架構	28
3.3	時序命名實體辨識	30
3.4	命名實體正規化	33
3.4.1	日期與時間正規化	33
3.4.2	期間正規化	34
3.4.3	頻率正規化	35
3.5	資料後處理Post-Processing	35
3.6	頻率轉換	37
3.7	問卷調查之研究架構與假說	38
第四章	研究結果與討論	40
4.1	實驗設計	40
4.1.1	資料來源	40
4.2	評估方式	40
4.3	實驗結果與錯誤分析	41
4.3.1	實驗結果	41
4.3.2	錯誤分析	42
4.4	問卷調查資料分析	43
4.4.1	雛型系統畫面	44
4.4.2	描述性資料統計結果	45
4.4.3	信效度分析	46
4.4.4	研究假說檢定	47
第五章	結論與未來發展	51
5.1	結論	51
5.2	研究貢獻與管理意涵	51
5.3	未來發展	52
參考文獻	53
附件一 問卷	56

圖目錄
圖1- 1 醫療資訊系統範疇 1
圖1- 2 論文架構 8
圖2- 1 住院病摘格式架構 17
圖2- 2 住院病摘格式架構中的ENTITY SET集合 17
圖2- 3 命名實體正規化範例 18
圖2- 4 DAVIS(1989)科技接受模式 21
圖2- 5 DELONE & MCLEAN(2003)資訊系統成功模式 24
圖3- 1系統發展方法的研究生命週期循環圖 26
圖3- 2系統發展研究方法論流程圖 27
圖3- 3 系統架構 29
圖3- 4 BCT NER系統架構 32
圖3- 5資料後處理流程 36
圖4- 1 範例一標註前	44
圖4- 2範例一標註後	44
圖4- 3範例二標註前	45
圖4- 4範例二標註後	45
圖4- 5假說檢定結果	48
圖4- 6 SMARTPLS執行結果	50

表目錄
表1- 1時間性特徵	5
表1- 2頻率表(資料來源:健保署藥品給藥頻率)	6
表2- 1 ISO 8601 10
表2- 2正規表示式常見語法 11
表2- 3命名實體分類 13
表3- 1本研究NER使用技術	31
表3- 2 頻率辨識RULE-BASED清單	33
表3- 3日期與時間正規表示式	34
表3- 4 期間正規表示式	34
表3- 5 頻率辨識PATTERN與標準化結果對照表	35
表3- 6 頻率設定資料表	37
表3- 7 本研究資管理論研究架構	38
表4- 1時序敘述類型分布	40
表4- 2評估結果(調整前)	41
表4- 3評估結果(調整後)	42
表4- 4問卷敘述性統計	46
表4- 5 信效度分析	47
表4- 6研究變數之區別效度	47
表4- 7 路徑分析結果	48
表4- 8 假說檢定結果	49

1.	Berger, A. L., Pietra, V. J. D., & Pietra, S. A. D. (1996). A maximum entropy approach to natural language processing. Comput. Linguist., 22(1), 39-71. 
2.	Buzhou Tang, Yonghui Wu, Min Jiang, Yukun Chen, Joshua C Denny,Hua Xu. (2013). A hybrid system for temporal information extraction from clinical text. Jamia, 20(5), 828-835. 
3.	Chang, A. X., & Manning, C. D. (2012). SUTime: A library for recognizing and normalizing time expressions. Paper presented at the Lrec, 3735-3740. 
4.	Chinchor, N. A. (1998). {Overview of MUC-7/MET-2}. Paper presented at the Proceedings of the Seventh Message Understanding Conference (MUC-7), , 1 doi:citeulike-article-id:5416136
5.	Chinchor, N. (1995). Four scorers and seven years ago: The scoring method for MUC-6. Paper presented at the Proceedings of the 6th Conference on Message Understanding, Columbia, Maryland. 33-38. doi:10.3115/1072399.1072403
6.	Chris Manning, S. U.Stanford CoreNLP. Retrieved from http://nlp.stanford.edu/software/corenlp.shtml
7.	Fisher, D., Soderland, S., Feng, F., & Lehnert, W. (1995). Description of the UMass system as used for MUC-6. Paper presented at the Proceedings of the 6th Conference on Message Understanding, Columbia, Maryland. 127-140. doi:10.3115/1072399.1072412
8.	ISO 8601:Representations of dates and times (1988). 
9.	Ji, H., & Grishman, R. (2011). Knowledge base population: Successful approaches and challenges. Paper presented at the Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies - Volume 1, Portland, Oregon. 1148-1158. 
10.	Kim, J. D., Ohta, T., Tateisi, Y., & Tsujii, J. (2003). GENIA corpus--semantically annotated corpus for bio-textmining. Bioinformatics, 19 Suppl 1, i180-2. 
11.	Lafferty, J. D., McCallum, A., & Pereira, F. C. N. (2001). Conditional random fields: Probabilistic models for segmenting and labeling sequence data. Paper presented at the Proceedings of the Eighteenth International Conference on Machine Learning, 282-289. 
12.	Leitner, F., Mardis, S. A., Krallinger, M., Cesareni, G., Hirschman, L. A., & Valencia, A. (2010). An overview of BioCreative II.5. IEEE/ACM Trans. Comput. Biol. Bioinformatics, 7(3), 385-399. doi:10.1109/tcbb.2010.61
13.	Lu, Z., Kao, H. Y., Wei, C. H., Huang, M., Liu, J., Kuo, C. J., . . . Wilbur, W. J. (2011). The gene normalization task in BioCreative III. BMC Bioinformatics, 12 Suppl 8, S2. doi:10.1186/1471-2105-12-S8-S2
14.	McCallum, A., & Li, W. (2003). Early results for named entity recognition with conditional random fields, feature induction and web-enhanced lexicons. Paper presented at the Proceedings of the Seventh Conference on Natural Language Learning at HLT-NAACL 2003 - Volume 4, Edmonton, Canada. 188-191. doi:10.3115/1119176.1119206
15.	Mikheev, A., Moens, M., & Grover, C. (1999). Named entity recognition without gazetteers. Paper presented at the Proceedings of the Ninth Conference on European Chapter of the Association for Computational Linguistics, Bergen, Norway. 1-8. doi:10.3115/977035.977037
16.	Mizuki Morita, Yoshinobu Kano, Tomoko Ohkuma, Mai Miyabe,Eiji Aramaki. (2013). Overview of the NTCIR-10 MedNLP task. Paper presented at the Proceedings of the NTCIR-10 Conference, Tokyo. 
17.	Morgan, A., Lu, Z., Wang, S. X., Cohen, A., Fluck, J., Ruch, P., . . . Hirschman, L. (2008). Overview of BioCreative II gene normalization. GenomeBiology.Com, 9(SUPPL. 2) doi:10.1186/gb-2008-9-s2-s3
18.	Nunamaker, J. F., Jr., & Chen, M. (1990). Systems development in information systems research. Paper presented at the System Sciences, 1990., Proceedings of the Twenty-Third Annual Hawaii International Conference on, , iii 631-640 vol.3. doi:10.1109/HICSS.1990.205401
19.	Sang, E. F. T. K. (2000). Text chunking by system combination. Paper presented at the Proceedings of the 2nd Workshop on Learning Language in Logic and the 4th Conference on Computational Natural Language Learning - Volume 7, Lisbon, Portugal. 151-153. doi:10.3115/1117601.1117638
20.	Sang, E. F. T. K. (2002). Introduction to the CoNLL-2002 shared task: Language-independent named entity recognition. Paper presented at the Proceedings of the 6th Conference on Natural Language Learning - Volume 20, 1-4. doi:10.3115/1118853.1118877
21.	Sang, E. F. T. K., & Meulder, F. D. (2003). Introduction to the CoNLL-2003 shared task: Language-independent named entity recognition. Paper presented at the Proceedings of the Seventh Conference on Natural Language Learning at HLT-NAACL 2003 - Volume 4, Edmonton, Canada. 142-147. doi:10.3115/1119176.1119195
22.	Uzuner, O., Solti, I., & Cadag, E. (2010). Extracting medication information from clinical text. Jamia, 17(5), 514-518. 
23.	Verhagen, M., Gaizauskas, R., Schilder, F., Hepple, M., Katz, G., & Pustejovsky, J. (2007). SemEval-2007 task 15: TempEval temporal relation identification. Paper presented at the Proceedings of the 4th International Workshop on Semantic Evaluations, Prague, Czech Republic. 75-80. 
24.	Verhagen, M., Saur, R., #237, Caselli, T., & Pustejovsky, J. (2010). SemEval-2010 task 13: TempEval-2. Paper presented at the Proceedings of the 5th International Workshop on Semantic Evaluation, Los Angeles, California. 57-62. 
25.	方鶯珍, & 璩大成. (2004). 實證醫學簡介. 北市醫學雜誌, 1(2), 233-237. doi:10.6200/TCMJ.2004.1.2.16
26.	林芬慧(Fen-Hui Lin). (2001). 調查研究法在資訊管理研究之使用－以“資訊管理年會論文集”之論文內容討論. 資訊管理學報, 8(1), 81-101. 
27.	梁定澎. (1997). 資訊管理研究方法總論. 資訊管理學報, 4(1), 1-6. 
28.	蔡宗宏(Chung-Hung Tsai), & 黃暉庭(Hui-Ting Huang). (2007). 醫療資訊系統成功模型之研究. 醫務管理期刊, 8(4), 281-300.