現代人靜態的生活模式已然成為增加許多疾病風險的關鍵因素，尤其是過重與肥胖問題，著實可稱為現代文明病。已有研究證實，體重過重情形的預防或改善與生活作息及日常活動習慣具高度相關，而當中也有許多研究建議以減少靜態活動的時間來逐步改變不健康的生活型態，臨床上日常活動量亦為醫師進行健康評估的重要參考資訊。因此，本論文之研究目的為開發一個日常活動模式之辨識系統，隨身收集使用者活動資訊，並協助日常活動習慣分析與探討體能活動量，進而應用於改善日常生活中運動量不足的問題。
本研究方法為利用智慧型手機收集加速度訊號與位置移動資訊加以分析，並於伺服器端的自主健康管理平台開發辨識模型與活動資訊回饋，建構出活動模式辨識系統。辨識模型的建置主要是基於支持向量機的多類別分類器以及位置興趣點萃取方法所開發而成。經過十折交叉驗證，本論文所實作之動作辨識系統的準確率可達98.99%；實際生活環境使用測試後，活動模式辨識的準確率為91.12%。
本論文設計一個小樣本的臨床介入實驗，評估本活動模式辨識系統與自主健康管理平台的介入，是否可達到改善體重過重者日常活動量的行為改變成效。初步臨床實驗結果顯示，在整體活動量方面，實驗組與對照組於活動量前後測之變化量具有顯著差異，顯示本系統的介入對於提升整體體能活動量具有成效。
總結而言，本研究開發一個活動模式辨識系統並自動產生個人化的活動模式圖表，有效提供使用者活動紀錄與分析資訊，並且以實際臨床實驗驗證將此勸誘科技應用於體重過重者之體能活動量改善的可行性。

More and more people have an increased risk for various diseases due to having a sedentary lifestyle and being overweight. The relationship between physical activity and overweight has been brought to light. Several studies have suggested that breaking up sedentary time with periods of movement may help mitigate the unhealthy effects inherent in a sedentary lifestyle. The purpose of this research is to develop a daily activity pattern recognition system that collects activity information portably and assists in discovering and assessing activity habits and assessing physical activity levels, further to improve the physical activity.
For the purpose of this research, a system based on an analysis of accelerometer signals and location information collected by a smartphone was developed, and a server-side platform was developed to recognize activity patterns and provide personal activity journals. The recognition model was performed by Support Vector Machine and point of interest locations extraction. We evaluated the activity pattern recognition system with cross-validation and got an accuracy of 98.99%. The accuracy rate of real-life testing data was 91.12%.
We designed a small sample clinical trial to assess the behavior change in overweight people after the intervention of the Automatic Activity Pattern Recognition System and Wellness Self-management Platform. The results of this trial demonstrated the outcomes for behavior change after intervention.
To conclude, the system we proposed is efficient with regard to recognizing activities and generating personal activity patterns. The clinical trials represented the feasibility of applying this persuasive technology toward improving the physical activity of overweight people.

References
[1] Cause of Death Statistics. Available: http://www.doh.gov.tw/CHT2006/DM/DM2_2.aspx?now_fod_list_no=9510&class_no=440&level_no=1
[2] International Physical Activity Questionnaires. Available: http://www.ipaq.ki.se/ipaq.htm
[3] N. Bidargaddi, A. Sarela, L. Klingbeil, and M. Karunanithi, "Detecting walking activity in cardiac rehabilitation by using accelerometer," in 3rd International Conference on Intelligent Sensors, Sensor Networks and Information, ISSNIP, 2007, pp. 555-560.
[4] S. N. Blair and T. S. Church, "The Fitness, Obesity, and Health Equation," JAMA: The Journal of the American Medical Association, vol. 292, pp. 1232-1234, September 8 2004.
[5] H. Blunck, T. Godsk, K. Grønbæk, M. B. Kjærgaard, J. L. Jensen, T. Scharling, K. R. Schougaard, and T. Toftkjær, "PerPos: a platform providing cloud services for pervasive positioning," presented at the Proceedings of the 1st International Conference and Exhibition on Computing for Geospatial Research & Application, Washington, D.C., 2010.
[6] J. S. I. Caros, O. Chételat, P. Celka, and S. Dasen, "Very Low Complexity Algorithm for Ambulatory Activity Classification," presented at the European Medical & Biological Engineering Conference and IFMBE European Conference on Biomedical Engineering, 2005.
[7] C.-C. Chang and C.-J. Lin. (2001). LIBSVM: a library for support vector machines. Available: http://www.csie.ntu.edu.tw/~cjlin/libsvm
[8] P. Dollar, V. Rabaud, G. Cottrell, and S. Belongie, "Behavior recognition via sparse spatio-temporal features," in 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance 2005, pp. 65-72.
[9] D. Figo, P. C. Diniz, D. R. Ferreira, and J. M. P. Cardoso, "Preprocessing techniques for context recognition from accelerometer data," Personal Ubiquitous Comput., vol. 14, pp. 645-662, 2010.
[10] E. S. Ford, W. H. Giles, and A. H. Mokdad, "Increasing Prevalence of the Metabolic Syndrome Among U.S. Adults," Diabetes Care, vol. 27, pp. 2444-2449, October 1 2004.
[11] T. Fu, T. Wen, P. Yeh, and H. Chang, "Costs of metabolic syndrome-related diseases induced by obesity in Taiwan," Obesity Reviews, vol. 9, pp. 68-73, 2008.
[12] S. M. Grundy, J. I. Cleeman, S. R. Daniels, K. A. Donato, R. H. Eckel, B. A. Franklin, D. J. Gordon, R. M. Krauss, P. J. Savage, S. C. J. Smith, J. A. Spertus, F. Costa, and E. Summary, "Diagnosis and management of the metabolic syndrome: An American Heart Association/National Heart, Lung, and Blood Institute scientific statement," Current Opinion in Cardiology, vol. 21, pp. 1-6, 2006.
[13] G. N. Healy, D. W. Dunstan, J. Salmon, E. Cerin, J. E. Shaw, P. Z. Zimmet, and N. Owen, "Breaks in Sedentary Time," Diabetes Care, vol. 31, pp. 661-666, April 2008.
[14] V. Heyward, Advanced Fitness Assessment and Exercise Prescription, 6 ed.: Human Kinetics, 2010.
[15] J. M. Hollander and J. I. Mechanick, "Complementary and Alternative Medicine and the Management of the Metabolic Syndrome," Journal of the American Dietetic Association, vol. 108, pp. 495-509, 2008.
[16] C.-L. Huang, "Motion Vector Estimation based on RFID Localization and Video Sequence," PhD, Electrical Engineering, National Cheng Kung University, 2008.
[17] K. C. Huang, "Obesity and its related diseases in Taiwan," Obesity Reviews, vol. 9, pp. 32-34, 2008.
[18] T. Huynh, M. Fritz, and B. Schiele, "Discovery of activity patterns using topic models," presented at the Proceedings of the 10th international conference on Ubiquitous computing, Seoul, Korea, 2008.
[19] J. L. Johnson, C. A. Slentz, J. A. Houmard, G. P. Samsa, B. D. Duscha, L. B. Aiken, J. S. McCartney, C. J. Tanner, and W. E. Kraus, "Exercise training amount and intensity effects on metabolic syndrome (from Studies of a Targeted Risk Reduction Intervention through Defined Exercise)," Am J Cardiol, vol. 100, pp. 1759-66, Dec 15 2007.
[20] D. M. Karantonis, M. R. Narayanan, M. Mathie, N. H. Lovell, and B. G. Celler, "Implementation of a real-time human movement classifier using a triaxial accelerometer for ambulatory monitoring," IEEE Transactions on Information Technology in Biomedicine, vol. 10, pp. 156-167, 2006.
[21] A. M. Khan, Y. K. Lee, and S. Y. Lee, "Accelerometer's position free human activity recognition using a hierarchical recognition model," in 12th IEEE International Conference on e-Health Networking Applications and Services (Healthcom), 2010, pp. 296-301.
[22] A. M. Khan, Y. K. Lee, S. Y. Lee, and T. S. Kim, "Human Activity Recognition via an Accelerometer-Enabled-Smartphone Using Kernel Discriminant Analysis," in 5th International Conference on Future Information Technology (FutureTech), 2010, pp. 1-6.
[23] A. M. Khan, L. Young-Koo, S. Y. Lee, and K. Tae-Seong, "A Triaxial Accelerometer-Based Physical-Activity Recognition via Augmented-Signal Features and a Hierarchical Recognizer," IEEE Transactions on Information Technology in Biomedicine, vol. 14, pp. 1166-1172, 2010.
[24] T. A. Lakka and D. E. Laaksonen, "Physical activity in prevention and treatment of the metabolic syndrome," Applied Physiology, Nutrition, and Metabolism, vol. 32, pp. 76-88, 2007.
[25] C.-C. Lin, C.-S. Liu, M.-M. Lai, C.-I. Li, C.-C. Chen, P.-C. Chang, W.-Y. Lin, Y.-D. Lee, T. Lin, and T.-C. Li, "Metabolic syndrome in a Taiwanese metropolitan adult population," BMC Public Health, vol. 7, p. 239, 2007.
[26] C. Lombriser, N. B. Bharatula, D. Roggen, and G. Troster, "On-body activity recognition in a dynamic sensor network," presented at the Proceedings of the ICST 2nd international conference on Body area networks, Florence, Italy, 2007.
[27] C.-D. Lu, "Human behavior analysis with interaction and event detection from video streams," PhD, Electrical Engineering, National Cheng Kung University, 2008.
[28] Z.-C. Luo, "The development of activity recognition system using the smartphone with accelerometer," Master, Institute of Medical Informatics, National Cheng Kung University, 2010.
[29] J. Macqueen, "Some methods for classification and analysis of multivariate observations," Proceedings of the fifth Berkeley symposium on mathematical statistics and probability, vol. 1, pp. 281-297 http://ci.nii.ac.jp/naid/10019277817/en/, 1967.
[30] S. McKeever, J. Ye, L. Coyle, C. Bleakley, and S. Dobson, "Activity recognition using temporal evidence theory," Journal of Ambient Intelligence and Smart Environments, vol. 2, pp. 253-269, 2010.
[31] H. Mizuno, H. Nagai, K. Sasaki, H. Hosaka, C. Sugimoto, K. Khalil, and S. Tatsuta, "Wearable Sensor System for Human Behavior Recognition (First Report: Basic Architecture and Behavior Prediction Method)," in International Solid-State Sensors, Actuators and Microsystems Conference, TRANSDUCERS, 2007, pp. 435-438.
[32] T. Nakajima, V. Lehdonvirta, E. Tokunaga, and H. Kimura, "Reflecting human behavior to motivate desirable lifestyle," presented at the Proceedings of the 7th ACM conference on Designing interactive systems, Cape Town, South Africa, 2008.
[33] F. Naya, R. Ohmura, F. Takayanagi, H. Noma, and K. Kogure, "Workers' Routine Activity Recognition using Body Movements and Location Information," in 10th IEEE International Symposium on Wearable Computers, 2006, pp. 105-108.
[34] R. Nishkam, D. Nikhil, M. Preetham, and L. L. Michael, "Activity recognition from accelerometer data," ed: AAAI Press, 2005.
[35] D. Riboni and C. Bettini, "Context-Aware Activity Recognition through a Combination of Ontological and Statistical Reasoning," Ubiquitous Intelligence and Computing, vol. 5585, pp. 39-53, 2009.
[36] T. N. Robinson, "TELEVISION VIEWING AND CHILDHOOD OBESITY," Pediatric Clinics of North America, vol. 48, pp. 1017-1025, 2001.
[37] P. C. Roy, S. Giroux, B. Bouchard, A. Bouzouane, C. Phua, A. Tolstikov, and J. Biswas, Possibilistic Behavior Recognition in Smart Homes for Cognitive Assistance, 2010.
[38] W. Shuangquan, Y. Jie, C. Ningjiang, C. Xin, and Z. Qinfeng, "Human activity recognition with user-free accelerometers in the sensor networks," in International Conference on Neural Networks and Brain, ICNN&B 2005, pp. 1212-1217.
[39] M. Stikic, T. Huynh, K. Van Laerhoven, and B. Schiele, "ADL recognition based on the combination of RFID and accelerometer sensing," in Second International Conference on Pervasive Computing Technologies for Healthcare, PervasiveHealth 2008, pp. 258-263.
[40] M. Tsiros, N. Sinn, A. Coates, P. Howe, and J. Buckley, "Treatment of adolescent overweight and obesity," European Journal of Pediatrics, vol. 167, pp. 9-16, 2008.
[41] H. Zhenyu, J. Lianwen, Z. Lixin, and H. Jiancheng, "Gesture recognition based on 3D accelerometer for cell phones interaction," in IEEE Asia Pacific Conference on Circuits and Systems, APCCAS 2008, pp. 217-220.