在全球人口老化的趨勢之下，世界各國都面臨到人口老化問題。醫療跟社會福利制度上的不足。無法趕上社會人老化的速度。其中在面對老人的居家環境，面臨到很大的問題。大部分的老人比較希望採取居家養老的方式，只有少部分人願意前往安養中心等機構。而當今台灣家庭情形大多是雙薪家庭，無法分出人手來照顧。再加上長照護員、醫療人員、社工人員等的人力嚴重不足。使的居家照護變的困難重重。而隨著5G科技的發展物聯網開始被大家使用，居家感應器也開始被大量運用。家中各式各樣的物品也開始可以連上網路，像是智慧空調、智慧管家等物聯網除了幫助人們生活更便利之外，或許也能提供居家照護的服務。藉此解決人力不足的問題，在家中安裝感應器，偵測居住者的情形。也不會侵犯到隱私權。
本研究以此想法為出發點，探討居家感應器在居家養老上面的可行性，期望藉此能解決照護人力不足的情形。將各種類型的感應器裝設在家中，例如紅外線感應器、壓力感應器、水流感應器、瓦斯感應器、光線感應器等，利用不同種類的感應器來偵測收集家中的活動數據，將這些數據分類後並使用機器學習的演算法方式來進行資料的分析，期望能藉由此方式建構出一套行為分析模式。就此判斷受照顧者是否需要協助，如此一來即可達到遠端監控，藉此來解決居家照護人力不足的問題。

Under the trend of global population aging, countries around the world are facing an ageing population. Insufficient medical and social welfare systems. Can not catch up with the speed of aging people in society. Among them, facing the elderly's home environment, faced a big problem. Most of the elderly prefer to take the approach of home-based care, and only a small number of people are willing to go to institutions such as the Anyang Center. Most of the family situations in Taiwan today are double-income families, and they cannot be separated from the hands of the people. In addition, the long-term care workers, medical personnel, social workers and other personnel are seriously inadequate. It is difficult to make home care. With the development of 5G technology, the Internet of Things has begun to be used by everyone, and home sensors have begun to be used in large numbers. A wide variety of items in the home can also be connected to the Internet. Internet of Things such as smart air conditioners and smart housekeepers can provide home care services in addition to helping people to live more conveniently. In this way, the problem of insufficient manpower is solved, and sensors are installed at home to detect the situation of the occupants. It will not infringe on privacy.
   This study takes this idea as a starting point to explore the feasibility of home sensors in home care, and hopes to solve the shortage of care. Various types of sensors are installed in the home, such as infrared sensors, pressure sensors, water sensors, gas sensors, light sensors, etc., using different types of sensors to detect activity data collected in the home, these will be After the data is classified and the machine learning algorithm is used to analyze the data, it is expected that a behavior analysis mode can be constructed by this method. In this regard, it is judged whether the care recipient needs assistance, so that remote monitoring can be achieved, thereby solving the problem of insufficient manual care for the home.

目     錄
圖目錄................................................V
表目錄................................................V
第一章 緒論............................................1
1.1研究動機與背景.......................................1
  1.2研究目的..........................................2
第二章 文獻探討.........................................3
2.1政府長照計畫.........................................3
  2.2 9073養老模式......................................5
2.3相關論文文獻.........................................6
第三章 研究方法設計......................................8
3.1分析感應器資料解析行為模式.............................8
  3.2 實驗分析..........................................9
3.3找出各感應器之間的可取代性............................16
第四章 結論............................................18
參考文獻...............................................19
附錄英文論文............................................21
1. INTRODUCTION .......................................21
2. RELATED WORKS.......................................22
3. RESEARCH METHODS AND EXPERIMENTAL ANALYSIS..........23
4. CONCLUSION..........................................26
5. REFERENCE................................... .......26

                        圖目錄
圖1.1老年人口比率........................................1
圖3.1浴室資料..........................................14
圖3.2決策樹分析1.......................................14
圖3.3決策樹分析2.......................................15
圖3.4矩陣資料型態......................................16
                        表目錄
表2.1長照2.0方案.......................................3
表2.2長照人力不足.......................................5
表3.1數據資料..........................................9
表3.2數據資料..........................................10
表3.3數據資料..........................................11
表3.4廚房區域..........................................12
表3.5臥室區域..........................................12
表3.6浴室資料..........................................12
表3.7資料分割..........................................13
表3.8廚房區域..........................................15
表3.9 (燈光、門)的反應相似..............................16
表3.10各感應器狀態.....................................17

[1] Oscar D. Lara, and Miguel A. Labrador. "A survey on 
    Human Activity Recognition using Wearable Sensors," 
    IEEE Communications Surveys and Tutorials pp1192-
    1209, 2013 

[2] Juha Parkka, Miikka Ermes, Panu Korpipaa, Jani 
    Mantyjarvi, Johannes Peltola, and Ilkka Korhonen, 
    "Activity Classification Using Realistic Data From 
    Wearable Sensors," IEEE Transactions on information 
    technology in biomedicine, pp119-128, 2006 

[3] Jeff Donahue, Lisa Anne Hendricks, Sergio Guadarrama, 
    Marcus Rohrbach, Subhashini Venugopalan, Kate Saenko, 
    and Trevor Darrell” Long-term Recurrent Convolutional 
    Networks for Visual Recognition and Description,” In 
    Proceedings of the IEEE conference on computer vision 
    and pattern recognition, pp. 2625-2634, 2015

[4] Ahmad Jalal, Yeon-Ho Kim, Yong-Joong Kim, Shaharyar 
    Kamal, and Daijin Kim “Robust Human Activity 
    Recognition from Depth Video using Spatiotemporal 
    Multi-fused Features,” Pattern recognition, 61, 
    pp.295-308, 2017.

[5] Kent Larson, William J. Mitchell, Dr. Stephen S. 
    Intille, and Dr. Joseph A. Paradiso, “Activity 
    Recognition in the Home using Simple and Ubiquitous 
    Sensors,” In International conference on pervasive 
    computing (pp. 1-136). Springer, Berlin, 
    Heidelberg, April 2004,

[6] Tim van Kasteren, Athanasios Noulas,  Gwenn 
    Englebienne, and Ben Krose, “Accurate Activity 
    Recognition in a Home Setting.” In Proceedings of the 
    10th international conference on Ubiquitouscomputing, 
    pp. 1-9, ACM. September.2008

[7] Ordónez, F.J., De Toledo P.,and Sanchis, A.,“Activity 
    Recognition using Hybrid Generative/Discriminative 
    Models on Home Environments using Binary Sensors,” 
    Sensors, pp.5460-5477. 2013.

[8] Aiguo Wang, Guilin Chen, Cuijuan Shang, Miaofei 
    Zhang, and Li Liu, “Human Activity Recognition in a 
    Smart Home Environment with Stacked Denoising 
    Autoencoders,” In International Conference on Web-Age 
    Information Management pp. 29-40 , June 2016,

[9] C. D. Nugent, “Delivering care at home through a 
    smarter environment,” IEEE 5th International 
    Conference on Pervasive Computing Technologies for 
    Healthcare 
    (Pervasive Health) and Workshops, pp475-477, October 
    2011

[10] Guang Yang, and Knut Øvsthus, “The Challenges of the 
     IoT Solutions in a Home Care Project,” IEEE 
     International Conference on Computational Science 
     and Computational Intelligence (CSCI), pp1771-1774, 
     Dec. 2017

[11] D. V. Patil, and R. S. Bichkar, “A Hybrid 
     Evolutionary Approach To Construct Optimal Decision 
     Trees With Large Data Sets,” IEEE International 
     Conference on Industrial Technology, pp429-433, June 
     2007

[12] Hong-tao Yang, “Research on Cost Decision of 
     Specialized-Automobile Manufacturing Enterprise 
     Based on the Theory of Decision Tree,” IEEE 
     International Conference on Digital Manufacturing & 
     Automation, pp198-203, January 2011

[13] Adriana Albu, “From logical inference to decision 
     trees in medical diagnosis,”IEEE E-Health and 
     Bioengineering Conference (EHB), pp65-68, June 2017