目前骨盆底肌肉訓練療程，醫師除了透過問卷內容了解患者的復健狀況外，醫師在面對不同類型的患者時，醫師只能透過醫療經驗等針對該患者提供處置方式，不容易發現其他可能影響療程等隱藏因子，故無法作出適合該患者的最佳處置。
本論文主要目的在對尿失禁患者利用資料探勘中的決策樹演算法進行分類，期望能探討  不同類型的患者中影響尿失禁治療結果等因子，提供醫師更清楚瞭解患者復健狀況。當有新患者需要進行復健運動時，即可透過決策樹分類模組達到預測的作用，醫師能適時給予患者協助。實作結果表明，各分類模組均可充分將目前患者資料正確地分類至各群組，有效協助醫師應用於臨床看診。

Currently, doctors in addition to the questionnaire to understand urinary incontinence through rehabilitation situation, the doctor in the face of different types of patients only by his medical experience providing for the disposition of the patient. This way is not easy to find other treatments that may affect other hidden factor, it is not appropriate for the patient to make the best deal.
The purpose of this thesis is used decision tree of data mining to discuss factors which affect result of urinary incontinence treatment of different types of patients. Providing that doctor can understand patient’s situation clearly. When a new patient needs to do pelvic floor muscle training, we can predict the result of treatment by using classification model of decision tree. Doctor can give patients assistance in pelvic floor muscle training to improve the treatment effect. The experimental results show that our approach can accurately classify the data to help doctor diagnose patients.

目錄
第一章 緒論	1
1.1研究背景	1
1.2研究動機與目的	2
1.3論文架構	3
第二章 相關研究	4
2.1尿失禁簡介	4
2.2尿失禁治療方法	5
2.2.1非手術治療	5
2.2.2手術治療	7
2.3資料探勘	7
2.4決策樹演算法	10
2.4.1常見的決策樹演算法	10
2.4.2 C4.5決策樹演算法之定義	12
第三章 研究方法	19
3.1知識探勘流程	19
3.2尿失禁病症分析流程	21
3.2.1原始資料	24
3.2.2資料轉換及預處理	25
3.2.3選擇目標欄位	26
3.2.4資料取樣	27
3.2.5資料探勘演算法	28
3.2.6臨床醫師模型驗證	30
第四章 實作結果與分析	32
4.1實作環境說明	32
4.2資料類別分組	33
4.3決策樹結果分析	35
4.3.1決策樹模組群組	35
4.3.2分類模組正確率驗證	43
第五章 結論與未來方向	47
5.1結論	47
5.2未來方向	47
參考文獻	49
附錄一 Urogenital Distress Inventory(UDI-6)	51
附錄二 Incontinence Impact Questionnaire(IIQ-7)	52
附錄三 英文論文	53

圖目錄
圖2.1決策樹模組	10
圖2.2訊息熵值變化	16
圖2.3性別屬性作為決策樹節點	18
圖2.4購買房車決策樹模組	18
圖3.1知識探勘流程	20
圖3.2尿失禁病症分析流程圖	23
圖3.3用藥紀錄資料(具有空值資料欄位)	25
圖3.4訓練資料集	28
圖3.5測試資料集	31
圖4.1年齡區間分布	33
圖4.2復健次數區間分布	34
圖4.3決策樹模組-年齡50-59歲	35
圖4.4決策樹模組-年齡60-69歲	36
圖4.5決策樹模組-年齡70-79歲	37
圖4.6決策樹模組-復健次數16-30次	39
圖4.7決策樹模組-復健次數31次以上	40
圖4.8決策樹模組-SUI患者	41
圖4.9決策樹模組-OAB患者	42
圖4.10決策樹模組-Other患者	43
圖4.11混亂矩陣	44

表目錄
表2.1常見的決策樹演算法	11
表2.2觀察家庭是否購買房車的相關資料	15
表3.1目標欄位	27
表4.1實作環境規格	32
表4.2決策樹模組分類規則-年齡50-59歲	36
表4.3決策樹模組分類規則-年齡60-69歲	36
表4.4決策樹模組分類規則-年齡70-79歲	38
表4.5決策樹分類規則-復健次數16-30次	39
表4.6決策樹模組類規則-復健次數31次以上	40
表4.7決策樹模組分類規則-SUI患者	41
表4.8決策樹模組分類規則-OAB患者	42
表4.9決策樹分類規則-Other患者	43
表4.10測試資料驗證模組-年齡	44
表4.11測試資料驗證模組-復健次數	44
表4.12測試資料驗證模組-Type	45
表4.13分類規則-改善狀況良好	46

[1]Maryam Kashanian,  Shadab Shah Ali,  Mitra Nazemi,  Shohreh Bahasadri , "Evaluation of the effect of pelvic floor muscle training (PFMT or Kegel exercise) and assisted pelvic floor muscle training (APFMT) by a resistance device (Kegelmaster device) on the urinary incontinence in women comparison between them: a randomized trial,"  European Journal of Obstetrics & Gynecology and Reproductive Biology, vol. 159, no. 1, pp.218–223, 2011.
[2]Diokno A, Yuhico M., Jr Preference, "Compliance and initial outcome of therapeutic options chosen by female patients with urinary incontinence," J Urol, vol. 154, pp.1727–1730, 1995.
[3]Cam C, Sakali M, Ay P, Cam M, Karateke A, "Validation of the short forms of the incontinence impact questionnaire (IIQ-7) and urogenital distress inventory (UDI-6) in a Turkish population," Neurourol Urodyn, vol. 26, no. 11, pp.29–33,  2007. 
[4]Schwartz, M. S., Biofeedback: A practitioner's guide, 2nd  ed., New York: Guilford Press, pp. 908, 1995.
[5]Kegel AH., "Progressive resistance exercise in functional restoratio of the perineal muscles," Am Journal of Obstet Gynecol, vol. 56, no. 2, pp.238-48, 1948.
[6]Jocelyne Fayn, "A Classification Tree Approach for Cardiac Ischemia Detection Using Spatiotemporal Information From Three Standard ECG Leads," IEEE Transactions on Biomedical Engineering, vol. 58, no. 1, pp. 95-102, January 2011. 
[7]Junghwan Cho, Xiaopeng Li, Zhiyong Gu, and Pradeep U. Kurup, "Recognition of Explosive Precursors Using Nanowire Sensor Array and Decision Tree Learning," IEEE Sensors Journal, vol. 12, no. 7, pp.2384-2391, July 2012.
[8]Guleng Sheri, and David Corne, "Learning-Assisted Evolutionary Search for Scalable Function Optimization: LEM(ID3)," IEEE Congress on Evolutionary Computation, pp.1-8, July 2010.
[9]Leszek Rutkowski, Fellow, IEEE, Lena Pietruczuk, Piotr Duda, and Maciej Jaworski, "Decision Trees for Mining Data Streams Based on the McDiarmid’s Bound," IEEE Transactions on Knowledge and Data Engineering, vol. 25, no. 6, pp.1272-1279, June 2013.
[10]Timo Horeis, Bernhard Sick, "Collaborative Knowledge Discovery & Data Mining: From Knowledge to Experience," Proceedings of the 2007 IEEE Symposium on Computational Intelligence and Data Mining(CIDM 2007), pp.421-428, 2007.