由於體感技術的興起，使體感互動應用及體驗式學習逐漸發展為虛擬實境與情境式學習的主流趨勢，其衍生的市場價值與潛力受到產官學界的重視。然而，現有體感技術對於使用者肢體動作的判定準確率與即時性上，仍有改善空間。
本論文提出結合決策樹方法之體感技術，並以需複雜且細膩操作的高爾夫球運動體感互動系統為例，經過兩階段受試者進行系統驗證測試：第一階段30人，第二階段100人；確實可有效提升使用者肢體動作與體感互動系統動作輸出的準確率與提供即時的流暢操作體驗。決策樹演算體感技術實現於體感高爾夫互動系統可提高肢體動作對應體感系統動作辨識的準確率、流暢度與直覺式操作體驗進行意向問卷調查，以描述性統計分析藉以了解使用者對於本系統的滿意度程度。高爾夫球體感運動互動系統包含二特點：(1) 應用CyWee動作感知無線手把作為體感控制器，採開放式PC-BASE系統具有技術擴充性與多元應用發展；(2)運用決策樹方法，進行手把實際動作與3D數位人物動作即時性的比對判斷，以利提高肢體動作判定的準確率與流暢性，實現輸入介面與體感系統間之即時互動體驗。在學術理論上，成功實作整合機器學習方法及體感技術，具體提昇體感技術的準確率與流暢度。此外，在教學方法理論方面，可延伸應用於運動與技能培訓的創新教學方法，為未來虛擬實境結合體感技術的數位學習研究領域，奠定紮實理論與實務基礎。

Because of the rise of somatosensory technology, somatosensory interactive application and experiential learning has become the mainstream of virtual reality and situated learning gradually. Its additional market values and potentialities also increasingly draw attention from Industry-Government-Academy circles. However, the current somatosensory technology still needs to improve the correct rate of judge and the immediacy of Feeback on users’ body movement.
The purpose of this study is to combine the somatosensory technology with decision tree, take a complicated golf somatosensory interactive system for example, and there are two groups of participants for the test of system verification .There are thirty participants in the first section, and one hundred participants in the second section. The experimental results reveal that the correct-rate of movement output, the immediacy of Feeback on users’ body movement can be efficiently improved
After the exploitation of decision tree algorithm of golf somatosensory interactive system, a questionnaire was designed to investigate satisfaction of users. This questionnaire puts emphasis on accuracy of movement judge, fluency of operation, and the experience of intuitive operation of the somatosensory system. The results of participants’ satisfaction were analyzed by the descriptive statistics.
In this study, there are two features of constructing golf somatosensory interactive system combined with decision tree. First, CyWee motion-sensing wireless handle was employed as a motion-sensing controller. Adopting open PC-BASE System can have technology expandability and multi-applications. Second, through the judge between the wireless handle and 3D digital human movement with the decision tree is capable of increasing the accuracy and fluency and achieves the instant interaction of input interface and somatosensory system. In academic theory, machine learning method and somatosensory technology are well integrated, and the accuracy and fluency of somatosensory technology are concretely improved as well. Moreover, the findings can be applied to innovative teaching methods of sports and skills cultivation from the aspect of instructional method theory. Furthermore, these results lay the theoretical and practical foundation of virtual reality combined with somatosensory technology in the field of digital study.

目 錄
摘 要 iii
Abstract iv
誌 謝 vi
目 錄 viii
表 目 錄 x
圖 目 錄 xii
符 號 說 明 xv
詞 彙 縮 寫 xvi
一、 緒 論 1
1.1 研究背景 1
1.2 研究動機及目的 4
1.3 研究內容及研究流程 8
二、 文獻探討 11
2.1 體感技術應用探討 11
2.2 體感技術之發展 13
2.3 體感技術結合VR運用與技能教育的發展趨勢探討 17
2.4 決策樹 19
2.5 高爾夫球訓練體感系統 20
三、 研究方法 25
3.1 體感高爾夫互動系統開發 25
3.1.1 系統架構與開發流程 25
3.1.2 決策樹於動作感知辨識技術 28
3.1.3 決策樹於訊號傳輸 30
3.1.4 決策樹於訊號轉換高爾夫系統參數 32
3.1.5 高爾夫球體感動作訊號樣本資料庫建立 33
3.1.6 實體動作感知與高爾夫系統 35
3.1.7 體感高爾夫系統流程 38
3.1.8 UI設計與相關術語 39
3.1.9 體感高爾夫物理原理 46
3.1.10 系統相關參數設定 51
3.1.11 體感高爾夫場地設計 60
3.2 體感高爾夫統計分析 62
四、 結果與討論 65
4.1 系統驗證結果 65
4.1.1 體感高爾夫揮球定義 66
4.1.2 高爾夫系統規則說明 72
4.1.3 高爾夫系統UI 73
4.1.4 高爾夫競賽模式 75
4.1.5 決策樹導入差異 76
4.2 描述性分析結果 82
4.2.1 問卷回收與樣本結構分析 82
4.2.2 整體滿意度敘述性統計分析 87
4.2.3 分群滿意度敘述性統計分析 91
4.3 結論 103
4.4 未來研究方向 105
參考文獻 106
附錄一 體感系統肢體動作樣本比對準確率測試 111
附錄二 高爾夫球體感互動系統使用行為調查問卷 114
附錄三 相關學術著作 116

參考文獻
[1]財團法人資訊工業策進會，2019，台灣數位內容產業年鑑，經濟部工業局。
[2]B. Jaruzelski and K. Dehoff, 2008, "Beyond borders: the global innovation 1000", strategy+ business, vol. 53, no. Winter, pp. 52-69.
[3]行政院(編印)，2008，國家發展重點計畫。
[4]吳致維，林建仲，2009，"互動式電子白板在國小教學之探討"，生活科技教育月刊，42卷，頁 14-25。
[5]王全世，2000，"資訊科技融入教學之意義與內涵"，資訊與教育雜誌，51卷， 頁 23-31。
[6]J. M. Balzotti and L. B. McCool, 2016, "Using digital learning platforms to extend the flipped classroom", Business and Professional Communication Quarterly, vol. 79, no. 1, pp. 68-80.
[7]覃桂昌，2010，軍訓課程以數位學習平台教學接受度探討之研究-以樹德科技大學為例，樹德科技大學，資訊管理系碩士論文。
[8]滕春麗，2010，精熟學習策略配合數位化診斷系統對高工學生數學科學習成效之研究，政治大學，應用數學系碩士論文。
[9]何榮桂，2002，"台灣資訊教育的現況與發展—兼論資訊科技融入教學"，資訊與教育雜誌，87卷，頁22-48。
[10]張基成，王秋錳，2008，"台北市高職教師資訊科技融入教學之影響因素" Journal of Educational Practice and Research，21卷，頁 97-132。
[11]T. Jones and R. Paolucci, 1999, "Research framework and dimensions for evaluating the effectiveness of educational technology systems on learning outcomes", Journal of Research on Computing in Education, vol. 32, pp. 17-27.
[12]王曉璿，1999，"資訊科技融入各科教學探究"，菁莪季刊，10卷，頁 18-24。
[13]陳韻雯，2009，桃園縣國民小學教師使用互動式電子白板之調查研究，國立臺北教育大學，國民教育學系碩士論文。
[14]溫嘉榮，2003，"教師如何將資訊融入學科成為教學工具"，教育研究，105卷，頁 75-81論文。
[15]K. Fu, T. Hainey, and G. Baxter, 2016, "A systematic literature review to identify empirical evidence on the use of computer games in business education and training" in European Conference on Games Based Learning, p. 232: Academic Conferences International Limited.
[16]R. Wang, S. DeMaria Jr, A. Goldberg, and D. Katz, 2016, "A systematic review of serious games in training health care professionals", Simulation in Healthcare, vol. 11, no. 1, pp. 41-51.
[17]K. Squire and H. Jenkins, 2003, "Harnessing the power of games in education", Insight, vol. 3, no. 1, pp. 5-33.
[18]T. Sourmelis, A. Ioannou, and P. Zaphiris, 2017, "Massively Multiplayer Online Role Playing Games (MMORPGs) and the 21st century skills: A comprehensive research review from 2010 to 2016", Computers in Human Behavior, vol. 67, pp. 41-48.
[19]E. Barsom, M. Graafland, and M. Schijven, 2016, "Systematic review on the effectiveness of augmented reality applications in medical training," Surgical endoscopy, vol. 30, no. 10, pp. 4174-4183.
[20]N. E. Seymour et al., 2002, "Virtual reality training improves operating room performance: results of a randomized, double-blinded study", Annals of surgery, vol. 236, no. 4, p. 458.
[21]J. L. McGrath et al., 2018, "Using virtual reality simulation environments to assess competence for emergency medicine learners", Academic Emergency Medicine, vol. 25, no. 2, pp. 186-195.
[22]H. Kemis, N. Bruce, W. Ping, T. Antonio, L. B. Gook, and H. J. Lee, 2012, "Healthcare monitoring application in ubiquitous sensor network: Design and implementation based on pulse sensor with arduino", Information Science and Service Science and Data Mining (ISSDM), 2012 6th International Conference on New Trends, pp. 34-38: IEEE.
[23]S. Zulfiqar, R. Zhou, F. Asmi, and A. Yasin, 2018, "Using simulation system for collaborative learning to enhance learner’s performance", Cogent Education, no. just-accepted, p. 1424678.
[24]H. Rheingold, 1991, Virtual Reality: Exploring the Brave New Technologies. Simon & Schuster Adult Publishing Group.
[25]J. Steuer, 1992, "Defining virtual reality: Dimensions determining telepresence", Journal of communication, vol. 42, no. 4, pp. 73-93.
[26]D. R. Begault and L. J. Trejo, 2000, "3-D sound for virtual reality and multimedia".
[27]L. P. Berg and J. M. Vance, 2017, "Industry use of virtual reality in product design and manufacturing: a survey", Virtual Reality, vol. 21, no. 1, pp. 1-17.
[28]B. I. Edwards, K. S. Bielawski, R. Prada, and A. D. Cheok, 2018, "Haptic virtual reality and immersive learning for enhanced organic chemistry instruction",Virtual Reality, pp. 1-11.
[29]T. Sanders and P. Cairns, 2010, "Time perception, immersion and music in videogames",Proceedings of the 24th BCS interaction specialist group conference, pp. 160-167: British Computer Society.
[30]D. Ritter and M. Johnson, 1997, "Virtual titrator: a student-oriented instrument", Journal of chemical education, vol. 74, no. 1, p. 120.
[31]C. Whitson and J. Consoli, 2009, "Flow theory and student engagement".
[32]Y.-Y. Song and L. Ying, 2015, "Decision tree methods: applications for classification and prediction", Shanghai archives of psychiatry, vol. 27, no. 2, p. 130.
[33]S. Wang, C. Fan, C.-H. Hsu, Q. Sun, and F. Yang, 2016, "A vertical handoff method via self-selection decision tree for internet of vehicles, IEEE Systems Journal, vol. 10, no. 3, pp. 1183-1192.
[34]L. Breiman, 2017, Classification and regression trees. Routledge.
[35]S. Zhang, X. Li, M. Zong, X. Zhu, and R. Wang, 2018, "Efficient knn classification with different numbers of nearest neighbors", IEEE transactions on neural networks and learning systems, vol. 29, no. 5, pp. 1774-1785.
[36]J. Han, M. Kamber, and J. Pei, 2011, Data mining: concepts and techniques. Morgan Kaufmann.
[37]H. Parvin, M. MirnabiBaboli, and H. Alinejad-Rokny, 2015, "Proposing a classifier ensemble framework based on classifier selection and decision tree", Engineering Applications of Artificial Intelligence, vol. 37, pp. 34-42.
[38]G. Ke et al. , 2017, "Lightgbm: A highly efficient gradient boosting decision tree" Advances in Neural Information Processing Systems, pp. 3146-3154.
[39]B. Zhang, Z. Wei, J. Ren, Y. Cheng, and Z. Zheng, 2018, "An empirical study on predicting blood pressure using classification and regression trees", IEEE access, vol. 6, pp. 21758-21768.
[40]C.-l. Jan, 2018, "An effective financial statements fraud detection model for the sustainable development of financial markets: evidence from Taiwan", Sustainability, vol. 10, no. 2, p. 513.
[41]楊亨利，黃泓彰等，2015，"基於決策樹與二元語言模型的網路用語轉譯系統"，電子商務學報，17卷，頁25-48。
[42]H. Rahmani, A. Mahmood, D. Q. Huynh, and A. Mian, 2014, "Real time action recognition using histograms of depth gradients and random decision forests", IEEE winter conference on applications of computer vision, pp. 626-633: IEEE.
[43]H. Ghasemzadeh and R. J. I. T. o. I. I. Jafari, 2010, "Physical movement monitoring using body sensor networks: A phonological approach to construct spatial decision trees",vol. 7, no. 1, pp. 66-77.
[44]K. King, S. W. Yoon, N. C. Perkins, and K. Najafi, 2008, "Wireless MEMS inertial sensor system for golf swing dynamics", Sensors and Actuators A: Physical, vol. 141, no. 2, pp. 619-630, 2008/02/15.
[45]H. Ghasemzadeh, V. Loseu, E. Guenterberg, and R. Jafari, 2009, "Sport training using body sensor networks: a statistical approach to measure wrist rotation for golf swing," presented at the Proceedings of the Fourth International Conference on Body Area Networks, Los Angeles, California. Available: https://doi.org/10.4108/ICST.BODYNETS2009.6035
[46]P. Kelly, A. Healy, K. Moran, and N. E. O’Connor, 2010, "A virtual coaching environment for improving golf swing technique," presented at the Proceedings of the 2010 ACM workshop on Surreal media and virtual cloning, Firenze, Italy. Available: https://doi.org/10.1145/1878083.1878098
[47]R. Burchfield and S. Venkatesan, 2010, "A Framework for Golf Training Using Low-Cost Inertial Sensors", 2010 International Conference on Body Sensor Networks, pp. 267-272.
[48]D. J. Kooyman, D. A. James, and D. D. Rowlands, 2013, "A feedback system for the motor learning of skills in golf", Procedia Engineering, vol. 60, pp. 226-231.
[49]A. Umek, Y. Zhang, S. Tomažič, and A. Kos, 2017, "Suitability of strain gage sensors for integration into smart sport equipment: A golf club example", Sensors, vol. 17, no. 4, p. 916.
[50]Y. Tanaka, H. Kon, and H. Koike, 2018, "A real-time golf-swing training system using sonification and sound image localization," Proceedings of the 24th ACM Symposium on Virtual Reality Software and Technology, pp. 1-2.
[51]A. Ikeda, Y. Tanaka, D.-H. Hwang, H. Kon, and H. Koike, 2019, "Golf training system using sonification and virtual shadow", presented at the ACM SIGGRAPH 2019 Emerging Technologies, Los Angeles, California. Available: https://doi.org/10.1145/3305367.3327993
[52]邱柏鈞，2012，基於體感網路的高爾夫揮桿分析，國立交通大學，網路工程研究所碩士論文。
[53]黃怡箴，2013，iGolf : 一個基於體感網路與雲端服務之高爾夫球揮桿學習系統，國立交通大學網路工程研究所碩士論文。
[54]郭癸賓，黃世育，2011，"以 Wii Fit 分析高爾夫揮桿動作之可行性評估" 華人運動生物力學期刊，頁18-26。