近幾年隨著社會不斷變遷，人口出生率及死亡率亦逐年降低，「人口老化」這一現象是多數國家所面臨的情況。隨著老年人口的比例不斷增加，相伴隨的老人照護問題更是近年關注的焦點議題，在機構式、社區式及居家式照護服務中，居家照護是最被接受的方式，也較具可行性，在老人居家養老的過程中，其健康狀態、照護方式是最需注意的問題。由於行為代表健康，本論文擬針對居家老人日常生活行為分析，從智慧居家系統佈建大量感測器，以獲取老人日常活動行為中的各項數據，並將大量感測器資料經過資料探勘技術，作為辨識老人日常生活行為初步判斷，再透過動態規劃演算法能比對相似度，進而找出獨居老人日常生活行為，最後再使用字串搜尋演算法將動態規劃比對完成的樣式與資料探勘技術中的循序樣式交叉比對，找出老人真正的日常生活行為。

In recent years, with the continuous social changes, the birth and death rates have also been declining year by year. The phenomenon of "population aging" is what most countries are facing. As the proportion of the elderly population continues to increase, the issue of elderly care accompanying the issue is the focus of attention in recent years. Home-based care is the most accepted and feasible method in institutional, community and home care services , In the process of elderly home care, the state of their health and the way of care are the most important issues to be addressed. In this paper, according to the analysis of everyday activities of the elderly living in the home, a large number of sensors are deployed in the wisdom home system to obtain the daily activities of the elderly Data and a large number of sensor data through the data mining technology as a preliminary assessment of the daily activities of the elderly, and then through the dynamic programming algorithm can be compared to similarities, and then find out the daily living behavior of single elderly people, and finally use the word The string search algorithm crosses the Pattern completed by the dynamic programming and the Sequential Pattern in the data mining technology to find out the real daily activities of the elderly.

目錄
第一章、簡介	1
第二章、相關研究	7
2-1. 利用健康照護及智慧居家系統探勘人類活動行為	8
2-2. COM：一種在家庭健康監測系統中探勘和監測人類活動模式的方法	9
2-3. DEANN：運用數據包絡的醫療保健分析方法分析和人工神經網路預測器官接收人體健康狀態	9
2-4. 大數據分析：為醫療保健組織了解其功能和潛在收益	10
2-5. 大數據應用和使用案例	12
2-6. 醫療保健分析從數據到知識到醫療保健改進	13
2-7. 醫療數據分析	14
2-8. 相關研究總結	15
第三章、系統架構與假設	16
3-1. 智慧居家系統介紹	16
3-2. Sensor佈建的假設	19
3-3. 感知家電的假設	20
3-4. 感知設備的假設	24
3-5. Gateway的假設	26
3-6. Server及數據庫的假設	28
第四章、數據分析與行為辨識	31
4-1. 定義資料清理及過濾資料計算方式	31
4-2. 定義資料時間限制及空間限制	33
4-3. 智慧居家環境資料收集及資料前處理	34
4-4. 實作第一次時間限制	36
4-5. 實作空間限制	36
4-6. 整理過後資料作時間切割	37
4-7. 定義Sensor標記	38
4-8. 字元串流之表示	40
4-9. 設定人工定義的行為模式(Pattern)	41
4-10. 動態規劃演算法 [31]（Dynamic Programming Algorithm,DP）	42
4-11. 最長公共子序列問題 [31]（Longest-Common-SubSequence ,LCS）	42
4-12. Boyer Moore Algorithm[32][33]	43
4-13. Needleman-Wunsch Algorithm[34][35]	46
4-14. Needleman-Wunsch Algorithm平行運算法[36]	47
4-15. Needleman-Wunsch Algorithm公式與評分機制	49
4-16. 資料比對演算流程	53
4-17. 與人工定義字串模式比對流程	56
第五章、資料比對實驗結果與效能評估	59
5-1. 透過Needleman-Wunsch Algorithm找到行為模式實驗結果	59
5-2. 平行演算法效能評估	61
第六章、結論	64
參考文獻	65
附錄-英文論文	70

  

圖目錄
圖 1各國65歲以上人口占總人口比率	2
圖 2居家健康照護物聯網系統架構圖	16
圖 3智慧居家系統Sensor佈建圖	19
圖 4智慧臥房展示及功能	21
圖 5智慧情境控制系統	22
圖 6健康照護系統	24
圖 7多網閘道器系統架構圖	26
圖 8大數據資料處理流程	30
圖 9感測器資料表	32
圖 10透過時間切割程式定義事件並將資料作切割	36
圖 11透過空間切割程式定義事件並將資料作切割	37
圖 12透過時間切割程式將資料作切割	38
圖 13感測器資料標記	39
圖 14將數據透過資料轉換成資料標記	39
圖 15將標記註記上資料流	40
圖 16將標記註記資料流流程	40
圖 17定義行為模式	41
圖 18 Boyer-Moore字串比對-完全符合	45
圖 19 Boyer-Moore Algorithm字串比對：不符合	46
圖 20 模式比對平行運算機制	48
圖 21 Needleman-Wunsch演算法-以ACD及ABCD字串為例	51
圖 22 Needleman-Wunsch Algorithm比對：以第一天及第二天資料為例	53
圖 23 30天數據庫資料	54
圖 24 Needleman-Wunsch Algorithm比對：從第1天統計到第30天	55
圖 25浴室行為配對	56
圖 26運用Needleman-Wunsch Algorithm比對兩邊資料	57
圖 27運用Boyer-Moore Algorithm比對兩邊資料	57
圖 28依據表4的數據，將程式執行時間減掉資料傳送時間的平均值	62
圖 29依據表4的數據，將CPU時間的平均值用直線圖表示	63
 
 
表目錄
表 1.系統功能比較表	15
表 2.Boyer-Moore 模式最後出現位置計算結果	44
表 3.Boyer-Moore 字串比對計算過程	45
表 4.Needleman-Wunsch-以ACD及ABCD字串為例計算表	52
表 5.演算法比較表	57
表 6.Needleman-Wunsch Algorithm統計一覽表	60
表 7.Needleman-Wunsch程式執行平行運算時資料寫入時間數據	61
表 8.Needleman-Wunsch執行平行運算CPU執行時間比較	63

參考文獻
[1] 	衛生福利部統計處-長期照護統計，2017
[2] 	行政院勞委會網站，2017
[3] 	內政部統計處，「內政統計通報」1997-04-05
[4] 	內政部統計處，「中華民國94年老人狀況調查報告」，1996-08
[5] 	康健雜誌58期，「誰來照顧你的父母？」，2003-09-01
[6] 	王健，馬軍，王翔著。第三章健康行為-健康教育學。高等教育出版社，2006
[7] 	楊菊賢，健康行為的建立，醫學與哲學，2001-03
[8] 	J. W. Han, H. Cheng, D. Xin, X. Yan, “Frequent pattern mining: current status and future directions,” Springer Science+Business Media,  Vol. 15, Issue 1, pp. 55–86, January 2007.
[9] 	Y. C. Wang, L. A. Kung, W. Y. C. Wang, and C. G. Cegielski, ”An integrated big data analytics-enabled transformation model: Application to health care,” Elsevier B.V. All rights reserved., vol. 55, pp. 64-79, January 2018.
[10] 	R. Agrawal, and R. Srikant, ”Fast Algorithms for Mining Association Rules,” IBM Almaden Research Center, Proceedings of the 20th VLDB Conference Santiago, Chile, 1994.
[11] 	R. J. Bayardo Jr., ”Efficiency Mining Long Patterns from Databases,” IBM Almaden Research Center, SIGMOD Conference, 1998.
[12] 	N. R. Mabroukeh, and C. I. Ezeife, “A taxonomy of sequential pattern mining algorithms,” University of Windsor, ACM Computing Surveys, Vol. 43, Article No.3, November 2010.
[13] 	N. Tax, N. Sidorova, R. Haakma, Will M.P. van der Aalst, “Mining Local Process Models,” Journal Of Innovation In Digital Ecosystems, vol. 3, pp. 183-196, ISBN: 978-3-642-02787-1,December 2016.
[14] 	M. Abouelhoda, M. Ghanem, “String Mining in Bioinformatics,”    SpringLink, July 2009.
[15] 	D. Maier, “The Complexity of Some Problems on Subsequences and Supersequences,” Journal of the ACM (JACM), vol. 25, pp. 322-336, April 1978. 
[16] 	L. Bergroth, H. Hakonen, T. Raita, “A Survey of Longest Common Subsequence Algorithms,” IEEE Xplore,27-29 Sept. 2000, ISBN: 0-7695-0746-8, August 2002.
[17] 	R. I. Greenberg, “Bounds on the Number of Longest Common Subsequences,”  Cornell University Library, v2, Aug 2003.
[18] 	M. Kiwi, M. Loebl, J. Matoušek, “Expected length of the longest common subsequence for large alphabets,” Cornell University Library, v2, Aug 2003.
[19] 	 R. S. Boyer, J. S. Moore, “A fast string searching algorithm,” Magazine Communication of the ACM (JACM), vol. 20, pp. 762-772, Oct. 1977.
[20] 	設計與實作物聯網閘道器以支援智慧居家照護，林俊凱，2017-06
[21] 	物聯網與大數據技術支援下設計及實作居家健康照護之緊急事件偵測與行為辨識，江得瑋，2014-06
[22] 	J. W. Han, M. Kamber, J. Pei, “Data Mining Concepts and Techniques 3/e,” Morgan Kaufmann, ISBN: 0123814790, Jul. 2011.
[23] 	T. C. Havens, J. C. Bezdek, C. Leckie, and M. Palaniswami, “Fuzzy C-Means Algorithm For Very Large Data,” IEEE TRANSACTIONS ON FUZZY SYSTEMS, vol. 20, no. 6, pp.1130-1146, May 2012.
[24] 	A. Yassine, S. Singh, and A. Alamri, “Mining Human Activity Pattern From Smart Home Big Data For Health Care Applications,” IEEE Access, Vol. 5, pp. 13131-13141, June 2017.
[25] 	P. Rashidi, and D. J. Cook, “COM: A Method for Mining and Monitoring Human Activity Patterns in Home-Based Health Monitoring Systems,” ACM Transactions on Intelligent Systems and Technology (TIST), vol. 4, Article No. 64., September 2013. 
[26] 	N. Misiunas, A. Oztekin, Y. Chen, and K. Chandra, “DEANN: A healthcare analytic methodology of data envelopment analysis and artificial neural networks for the prediction of organ recipient functional status,” Elsevier Ltd. All rights reserved., vol. 58, pp.46-54, April 2015.
[27] 	Y. Wang,L. Kung, and T. A. Byrd, “Big data analytics: Understanding its capabilities and potential benefits for healthcare organizations,”  Elsevier Inc. All rights reserved., vol. 126, pp.3-13, February 2016.
[28] 	P. C. K. Hung, “Big Data Applications and Use Cases,” Springer International, ISBN: 978-3-319-30146-4, 2016.
[29] 	H. Yang, and E. K. Lee, “Healthcare Analytics From Data to Knowledge to Healthcare Improvement,”  John Wiley & Sons, Inc. All rights reserved, ISBN: 978-0-787-97301-8, August 2016.
[30] 	C. K. Reddy, and C. C. Aggarwal, “Healthcare Data Analytics,” Chapman and Hall/CRC, ISBN:978-1-482-23211-0, June 2015. 
[31] 	T. H. Cormen, C. E. Leiserson, R. L. Rivest, and C. Stein, “Introduction to Algorithms 3rd Edition,” MIT Press, ISBN: 978-0-262-03384-8, September 2009.
[32] 	R. S. Boyer, and J. S. Moore, "A Fast String Searching Algorithm,". Communications of the ACM, vol.20, pp.762-772, October 1977.
[33] 	J. Jaffray. "Visualizing Brute Force, KMP, and Boyer-Moore," Github, Nov  2015.
[34] 	S. B. Needleman, C. D. Wunsch, "A general method applicable to the search for similarities in the amino acid sequence of two proteins," Journal of Molecular Biology, vol.48, pp.443-453, March 1970.
[35] 	M. Abdelraouf, "Interactive Visualization of Needleman-Wunsch Algorithm in Javascript," Github, Sep 2015.
[36] 	T. Naveed, I. S. Siddiqui, and S. Ahmed, "Parallel Needleman-Wunsch Algorithm for grid," Department of Computer Sciences & Engineering, Bahria University Islamabad, Pakistan, pp.1-6, July 2005.