臺灣博碩士論文加值系統

由於高齡化社會以及年輕一輩長期外宿工作，導致長輩有極大的機會長期處於單獨在家的狀態，而獨居時一旦發生意外常常會導致憾事發生，而隨著智慧住宅的普及，我們也可以利用這一點更加強化智慧住宅的應用，因此本論文為建立一套為獨居老人所設計的行為辨識程序。
本論文將收集使用者在智慧住宅內生活的數據，如紅外線移動感測、溫度感測等，而使用者會先將初始數據紀錄上他們的行為活動，將這些數據格式處理後，將帶入本論文的程式，此程式會將數據經由SVM 演算法的處理後電腦將可以藉由先前的數據進而學習，之後就可以分析使用者往後生活的行為模式，藉此判斷使用者的行為是否出現變化。

Due to the aging society and the long-term work of the younger generation, the elders have great opportunities to stay alone at home for a long time. Once an accident occurs, it often leads to regrets when the elders living alone. With the popularity of smart homes, we can also use this to strengthen the application of smart homes. Therefore, this paper would establish a activity recognition program designed for the elderly living alone.
This paper will collect user data living in smart homes, such as infrared sensor, temperature sensor, etc. The user will first record their behavior on the initial data, and then process the data, which will be used the program of this paper. The data will be processed by SVM algorithm. The computer will be able to learn from the previous data, and then the computer can analyze the user''s future life behavior patterns. This is used to judge whether the user''s behavior has changed.

摘要 I
ABSTRACT II
誌謝 III
目錄 IV
圖目錄 V
表目錄 VII
第一章 緒論 1
1.1 研究動機 1
1.2 研究目的 1
第二章 文獻回顧 4
2.1 活動識別的步驟 5
2.1.1 建置 5
2.1.2 數據收集 6
2.1.3 分類 7
2.1.4 識別 8
2.1.5 預測 9
2.2 選擇演算法 9
2.2.1 單純貝葉斯 (Naive Bayes) 9
2.2.2 決策樹(Decision tree) 11
2.2.3 支撐向量機(SVM) 11
第三章 研究方法-SVM演算法介紹 14
3.1 Linear SVM 14
3.2 Dual SVM 17
3.3 Kernel SVM 21
3.4 Soft-Margin SVM 24
第四章 系統設計 26
第五章 結果與討論 43
5.1 數據前處裡 43
5.2 實驗結果 48
第六章 結論與未來發展 59
參考文獻 60

[1] https://www.ndc.gov.tw/Content_List.aspx?n=695E69E28C6AC7F3
[2] http://www.tada2002.org.tw/About/IsntDementia
[3] http://www.kmuh.org.tw/www/kmcj/data/10602/14.htm
[4] N. Krishan and D. Cook. Activity recognition on streaming sensor data, 2012.
[5] F. Foerster, M. Smeja, J. Fahrenberg, Detection of posture and motion by accelerometry: a validation study in ambulatory monitoring, Computers in Human Behavior 15 (1999) 571 – 583.
[6] J. Alon, V. Athitsos, Q. Yuan, S. Sclaroff, A unified framework for gesture recognition and spatiotemporal gesture segmentation, IEEE Transactions on Pattern Analysis and Machine Intelligence 31 (2008) 1685–1699.
[7] https://www.csie.ntu.edu.tw/~cjlin/libsvm/
[8] D. Cook, Learning setting-generalized activity models for smart spaces, IEEE Intelligent Systems (to appear).
[9] N. Ravi, N. Dandekar, P. Mysore, M. L. Littman, Activity recognition from accelerometer data, IAAI’05: Proceedings of the 17th conference on Innovative applications of artificial intelligence (2005) 1541–1546.
[10] N. Ravi, N. Dandekar, P. Mysore, M. L. Littman, Activity recognition from accelerometer data, IAAI’05: Proceedings of the 17th conference on Innovative applications of artificial intelligence (2005) 1541–1546.
[11] http://casas.wsu.edu/datasets/
[12] CASAS: A Smart Home in a Box Diane J. Cook, Aaron S. Crandall, Brian L. Thomas, and Narayanan C. Krishnan
[13] WILEY SERIES ON PARALLELAND DISTRIBUTED COMPUTING Series Editor: Albert Y. Zomaya
[14] https://reurl.cc/aoMXQ
[15] https://blog.csdn.net/zrh_CSDN/article/details/81007851
[16] https://blog.csdn.net/keepreder/article/details/47168383
[17] https://blog.csdn.net/qq_38734403/article/details/80442535
[18]林軒田教授的機器學習技法http://yt1.piee.pw/H9ML6