Responsive image
博碩士論文 etd-0620118-143652 詳細資訊
Title page for etd-0620118-143652
論文名稱
Title
基於資料導向改善居家環境中的活動識別
A data-driven approach to recognize human daily living activity
系所名稱
Department
畢業學年期
Year, semester
語文別
Language
學位類別
Degree
頁數
Number of pages
60
研究生
Author
指導教授
Advisor
召集委員
Convenor
口試委員
Advisory Committee
口試日期
Date of Exam
2018-06-25
繳交日期
Date of Submission
2018-07-20
關鍵字
Keywords
隨機森林、深度學習、活動辨識、機器學習、智慧環境
machine learning, activity recognition, smart environments, random forest, deep learning
統計
Statistics
本論文已被瀏覽 5947 次,被下載 0
The thesis/dissertation has been browsed 5947 times, has been downloaded 0 times.
中文摘要
識別日常生活行為主要用於醫療照護上,能夠改善並追蹤病患的生活行為是否有異常,可以有效地減少人力資源使用,而識別日常生活活動也逐漸成為智慧家庭(Smart home)的根本要件,但相較於使用在醫療照護上,應用在智慧家庭牽扯到了更多有關隱私安全的問題,因此該如何在不增加侵入式感測器(Invasive sensor)的情況下有效提升準確度變成重要的議題。現今倚靠知識導向的做法已經可以達到很高的準確度,但會衍生依靠人的主觀想法去建立知識的問題,由於每個人的生活型態會不同,因此產生的知識或許會造成反效果,且會讓事前工作變得繁雜,所以本研究希望以較客觀的數據來建立個別個人化的模型。在華盛頓州立大學自適應進階研究中心(WSU CASAS)的資料集當中,相較於以往的研究,都是將CASAS的資料集加上穿戴式裝置(Body-worn)的資料,來達到建立限制式的知識導向效果,但本研究認為穿戴式裝置會牽涉到更多隱私以及便利性的問題,因此只採用純非侵入式感測器資料對多住戶日常生活行為的原始資料做處理,並且將資料依照我們所設計的編碼方式,有別於以往設定固定長度為區間當作輸入資料的研究,本研究設計順序型的編碼可以讓資料以單筆為訓練資料的方式進行訓練與驗證,以克服時間序列資料在不同時段出現頻率不同的問題。本研究將使用三種模型,分為長短期記憶(Long Short-Term Memory)的Sequence to Sequence 模型、隨機森林以及支持向量機器來進行建模,期望能透過機器學習的能力降低對於知識的依賴,讓資料集的預測結果更為平衡且能夠預測出新的活動組合,且能夠維持且提高各個活動在整體的準確度、精確度、召回率以及F1-measure。
Abstract
Recognizing activities of daily living is mainly used in medical care, record and track patients' daily living. It can substantially reduce burden of human resource. And recognizing daily living activity has become the key point of Smart home. However, compared with medical care, Smart home requires more privacy. Therefore, the main problem is how to improve accuracy without equipping with any invasive sensor.
Using Knowledge-driven approach have been conducted well, but Knowledge-driven relies on subjective restrictions. And each individual resident has their own living style, subjective restrictions probably has reached the limitations and needs a lot of work to do in advance. Therefore, our research employees an approach that based on objective method to recognize human's daily activity.
The research data was collected form WSU CASAS. Previous researches took advantage of camera or body-worn sensors to make rules of restriction, but we figure that body-worn sensor involve more privacy and convenience issue. Therefore, we only used data collected from non-invasive sensor. The coding method we designed features the time sequence in each individual datum, which allows us to randomly draw training data from the whole population. And it will solve the problem that the frequency of occurrence of activities vary from time to time.
Our research employed three models, which are random forest, support vector machine and sequence to sequence model. We expect that machine learning won't require prior knowledge to have a better balance prediction result among all activities and have ability to predict new combination of activities, so do the accuracy, the precision, the recall and the F1-measure.
目次 Table of Contents
論文審定書 i
中文摘要 ii
英文摘要 iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章 緒論 1
1.1 研究背景與動機 1
1.2 研究宗旨與目標 2
1.3 研究貢獻 3
第二章 相關文獻 5
2.1 活動辨識方法類型 5
2.2 感測器設置 6
2.3 特徵萃取 7
2.4 建模方式 8
2.5 單人和多人 8
2.6 編碼方式 9
第三章 研究方法 10
3.1 研究流程 11
3.2 資料集 13
3.3 資料處理 14
3.4 編碼方式 16
3.5 研究設計 19
3.6 模型架構 20
3.7 評量方式 24
第四章 實驗評估 26
4.1 單人資料集 27
4.1.1 編碼方式 27
4.1.1.1 簡單編碼 27
4.1.1.2 簡單編碼-考量未知感測器 28
4.1.1.3 考慮順序及數值型感測器 30
4.1.2 樣本平衡 31
4.1.2.1 原始資料集 32
4.1.2.2 Synthetic Minority Over-sampling Technique 34
4.2 多人資料集 37
4.2.1 編碼方式 37
4.2.1.1 簡單編碼 37
4.2.1.2 簡單編碼-考量未知感測器 38
4.2.1.3 考慮順序及數值型感測器 39
4.2.2 樣本平衡 41
4.2.2.1 原始資料集 41
4.2.2.2 Synthetic Minority Over-sampling Technique 44
4.3 實驗結果 46
第五章 結論 47
5.1 結論 47
5.2 未來展望 48
第六章 參考文獻 49
參考文獻 References
1. Gayathri, K. S., K. S. Easwarakumar, and Susan Elias. "Probabilistic ontology based activity recognition in smart homes using Markov Logic Network." Knowledge-Based Systems 121 (2017): 173-184.

2. Roy, Nirmalya, Archan Misra, and Diane Cook. "Infrastructure-assisted smartphone-based ADL recognition in multi-inhabitant smart environments." Pervasive Computing and Communications (PerCom), 2013 IEEE International Conference on. IEEE, 2013.

3. Alam, Mohammad Arif Ul, et al. "CACE: Exploiting behavioral interactions for improved activity recognition in multi-inhabitant smart homes." Distributed Computing Systems (ICDCS), 2016 IEEE 36th International Conference on. IEEE, 2016.

4. Chen, Liming, et al. "Sensor-based activity recognition." IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews)42.6 (2012).

5. Bahdanau, Dzmitry, Kyunghyun Cho, and Yoshua Bengio. "Neural machine translation by jointly learning to align and translate." arXiv preprint arXiv:1409.0473 (2014).

6. Chen, Liming, Chris D. Nugent, and Hui Wang. "A knowledge-driven approach to activity recognition in smart homes." IEEE Transactions on Knowledge and Data Engineering 24.6 (2012).

7. Wongpatikaseree, Konlakorn, et al. "Activity recognition using context-aware infrastructure ontology in smart home domain." Knowledge, Information and Creativity Support Systems (KICSS), 2012 Seventh International Conference on. IEEE, 2012.

8. Manzi, Alessandro, et al. "Two-person activity recognition using skeleton data." IET Computer Vision (2017).

9. Clarkson, Brian, Kenji Mase, and Alex Pentland. "Recognizing user context via wearable sensors." Wearable Computers, The Fourth International Symposium on. IEEE, 2000.

10. Bao, Ling, and Stephen S. Intille. "Activity recognition from user-annotated acceleration data." International Conference on Pervasive Computing. Springer, Berlin, Heidelberg, 2004.

11. Bettadapura, Vinay, et al. "Augmenting bag-of-words: Data-driven discovery of temporal and structural information for activity recognition." Computer Vision and Pattern Recognition (CVPR), 2013 IEEE Conference on. IEEE, 2013.

12. Crispim-Junior, Carlos F., et al. "Semantic event fusion of different visual modality concepts for activity recognition." IEEE transactions on pattern analysis and machine intelligence 38.8 (2016): 1598-1611.

13. Wang, Liang, et al. "Recognizing multi-user activities using wearable sensors in a smart home." Pervasive and Mobile Computing 7.3 (2011).

14. Lara, Oscar D., and Miguel A. Labrador. "A survey on human activity recognition using wearable sensors." IEEE Communications Surveys and Tutorials 15.3 (2013): 1192-1209.

15. Dernbach, Stefan, et al. "Simple and complex activity recognition through smart phones." Intelligent Environments (IE), 2012 8th International Conference on. IEEE, 2012.

16. Győrbíró, Norbert, Ákos Fábián, and Gergely Hományi. "An activity recognition system for mobile phones." Mobile Networks and Applications 14.1 (2009): 82-91.

17. Riboni, Daniele, et al. "Smartfaber: Recognizing fine-grained abnormal behaviors for early detection of mild cognitive impairment." Artificial intelligence in medicine 67 (2016): 57-74.

18. Benmansour, Asma, Abdelhamid Bouchachia, and Mohammed Feham. "Multioccupant activity recognition in pervasive smart home environments." ACM Computing Surveys (CSUR) 48.3 (2016): 34.

19. Alam, Mohammad Arif Ul, and Nirmalya Roy. "GeSmart: A gestural activity recognition model for predicting behavioral health." Smart Computing (SMARTCOMP), 2014 International Conference on. IEEE, 2014.

20. Rabiner, Lawrence R. "A tutorial on hidden Markov models and selected applications in speech recognition." Proceedings of the IEEE 77.2 (1989): 257-286.

21. Forney, G. David. "The viterbi algorithm." Proceedings of the IEEE 61.3 (1973): 268-278.

22. Singla, Geetika, Diane J. Cook, and Maureen Schmitter-Edgecombe. "Recognizing independent and joint activities among multiple residents in smart environments." Journal of ambient intelligence and humanized computing 1.1 (2010).

23. Chen, Liming, Chris Nugent, and George Okeyo. "An ontology-based hybrid approach to activity modeling for smart homes." IEEE Transactions on human-machine systems 44.1 (2014): 92-105.

24. Sutskever, Ilya, Oriol Vinyals, and Quoc V. Le. "Sequence to sequence learning with neural networks." Advances in neural information processing systems. 2014.

25. Krishnan, Narayanan C., and Diane J. Cook. "Activity recognition on streaming sensor data." Pervasive and mobile computing 10 (2014).
電子全文 Fulltext
本電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
論文使用權限 Thesis access permission:自定論文開放時間 user define
開放時間 Available:
校內 Campus: 已公開 available
校外 Off-campus: 已公開 available


紙本論文 Printed copies
紙本論文的公開資訊在102學年度以後相對較為完整。如果需要查詢101學年度以前的紙本論文公開資訊,請聯繫圖資處紙本論文服務櫃台。如有不便之處敬請見諒。
開放時間 available 已公開 available

QR Code