臺灣博碩士論文加值系統

智慧型手機的銷售量年年成長，而相關的安全性議題也越來越重要。為了保護智慧型手機內的資料，目前現有的智慧型手機使用者識別機制有侵入式與非侵入式兩種。傳統的驗證機制（密碼鎖、圖形鎖）屬於侵入式識別機制。非侵入式識別機制則不需要驗證介面，而是從背景收集使用者行為進行驗證。目前已有數種研究提出非侵入式識別機制，但皆採用同一姿勢進行實驗，未考慮不同姿勢造成的影響。首先本研究對不同姿勢下收集到的行為資料分析，證實不同姿勢下的資料彼此之間有顯著差異。第二部分以應用的角度而言，若混和不同姿勢的資料建模、測試的實驗與各姿勢行為資料獨立建模、測試的實驗相比準確率沒有下降很多，則可以直接忽略姿勢影響，混和各姿勢的資料進行建模。此問題將以動態方法進行實驗並根據實驗結果告知後續研究者可以直接混和各姿勢的資料進行建模、測試。最後推薦可以避免姿勢影響且實驗效果最佳的分類器。

Smartphone sales obviously grew in this years, so the associated security issues about smartphone has become more important. In order to protect the data within the smartphone, intrusive and non-intrusive user authentication mechanisms were developed. Traditional authentication mechanisms like number lock and pattern lock are intrusive user authentication mechanism. Non-intrusive user authentication mechanism doesn’t require any user interface, but collect user’s behavior in the background and authenticate it. Several non-intrusive authentication mechanisms were proposed, but all of them collected user behavior in one fixed posture. These mechanisms didn’t take posture’s effect into consideration. First, for this study, we analyze user’s behavior data in different postures and confirm that user’s behavior data has significant differences in different postures. Second, from the view of the application, if the accuracy that use mixed posture behavior data’s model to predict isn’t significantly lower than the accuracy that separately use single posture behavior data’s model to predict, we can directly neglect posture’s effect and use mixed posture behavior data’s model to predict. This problem will be discussed by doing the experiment in dynamics-based approach and then inform the future researchers that they can use mixed posture behavior data’s model to predict according to the experiment result. Finally, we recommend the best classifier that can avoid the posture’s effect and have the best prediction accuracy.

中文摘要 i
Abstract ii
目錄 iii
圖目錄 vi
表目錄 vii
一、 緒論 1
1-1 研究背景 1
1-2 研究動機 3
1-3 論文架構 4
二、 文獻探討 5
2-1 相關文獻探討 5
2-1-1 以方位感測器為基礎建構而成的非侵入式識別機制 5
2-1-2 以觸控螢幕為基礎建構而成的非侵入式識別機制 9
2-2 分類器 10
2-2-1 Naive Bayes 10
2-2-2 K-Nearest Neighbors 10
2-2-3 RBFSVM 11
2-2-4 CART 11
2-2-5 Back-propagation Neural Network 11
三、 實驗設計 12
3-1 資料來源 12
3-2 識別特徵 15
3-2-1 動態方法 15
3-2-2 方位感測器相關識別特徵 16
3-2-3 觸控螢幕相關識別特徵 19
3-3 資料轉換及正規化 23
3-4 實驗假設 24
3-5 ANOVA變異數檢定實驗設計 24
3-6 識別機制實驗設計 25
3-6-1 以learning curve決定各分類器的訓練樣本量 25
3-6-2 識別機制實驗流程 26
3-6-3 抽樣方式 27
3-6-4 特徵選取 29
3-6-5 訓練分類器參數 32
四、 實驗結果 33
4-1 ANOVA變異數檢定實驗結果與分析 33
4-1-1 方位感測器相關特徵行為資料ANOVA變異數檢定 33
4-1-2 觸控螢幕相關特徵行為資料ANOVA變異數檢定 35
4-2 識別機制實驗結果與分析 38
4-2-1 方位感測器相關特徵識別機制實驗結果與分析 38
4-2-2 觸控螢幕相關特徵識別機制實驗結果與分析 43
4-2-3 方位感測器+觸控螢幕相關特徵識別機制實驗結果與分析 47
五、 結論及未來展望 50
5-1 結論 50
5-2 研究貢獻 50
5-3 未來展望 52
參考文獻 53
附錄一 方位感測器相關特徵 55
附錄二 觸控螢幕相關特徵 57
附錄三 觸控螢幕相關特徵計算方式 59
附錄四 Learning Curve實驗結果 60

〔1〕 Gartner®, “Gartner Says Smartphone Sales Grew 46.5 Percent in Second Quarter of 2013 and Exceeded Feature Phone Sales for First Time”, available at: http://www.gartner.com/newsroom/id/2573415 (accessed 8 January 2014), 2013.
〔2〕 Gartner®, “Gartner Says Smartphone Sales Accounted for 55 Percent of Overall Mobile Phone Sales in Third Quarter of 2013”, available at: http://www.gartner.com/newsroom/id/2573415 (accessed 8 January 2014), 2013.
〔3〕 Matthew Boyle, Avraham Klausner, David Starobinski, Ari Trachtenberg, Hongchang Wu, “Gait-based User Classification Using Phone Sensors”, pp. 1-11, July 2011.
〔4〕 D. Gafurov, K. Helkala, and T. Søndrol, “Biometric Gait Authentication Using Accelerometer Sensor,” Journal of Computers, vol. 1, pp.51-59, October/November 2006.
〔5〕 Allano, L., Morris, A.C., Sellahewa, H., Garcia-Salicetti, S., Koreman, J., Jassim, S., Ly-Van, B., Wu, D. & Dorizzi, B., “Non intrusive multi-biometrics on a mobile device: a comparison of fusion techniques”, Proc. SPIE Conference on Biometric Techniques for Human Identification III, pp. 1-12, Orlando, U.S., 2006
〔6〕 M. Conti, I. Z. Zlatea, and B. Crispo, “Mind how you answer me: transparently authenticating the user of a smartphone when answering or placing a call.” In Proceedings of the 6th ACM Symposium on Information, Computer, and Communications Security, (ASIACCS '11). ACM, New York, NY, USA. pp. 249-259, March 22–24, 2011.
〔7〕 許振揚，「非侵入式多模組之手機使用者識別機制：基於動態方法」，1-39頁，2012年10月
〔8〕 Tao Feng, Ziyi Liu, Kyeong-An Kwon, Weidong Shi, Bogdan Carbunar, Yifei Jiang, Nhung Nguyen, “Continuous Mobile Authentication using Touchscreen Gestures”, In Proceedings of the 12th IEEE Conference on Technologies for Homeland Security (HST), pp. 1-6, Waltham, MA, November 2012.
〔9〕 HTC, “new HTC One Specs”, available at: http://www.htc.com/tw/smartphones/htc-one/#/ (accessed 26 January 2014), 2013.
〔10〕 Google, “Jelly-Bean”, available at: http://developer.android.com/about/versions/jelly-bean.html (accessed 26 January 2014), 2013.
〔11〕 Matthias Böhmer, Brent Hecht, Johannes Schöning, Antonio Krüger, Gernot Bauer, “Falling Asleep with Angry Birds, Facebook and Kindle - A Large Scale Study on Mobile Application Usage”, Proceedings of the 13th International Conference on Human-Computer Interaction with Mobile Devices and Services, Stockholm, Sweden, August 2011.
〔12〕 Google, “Android Document: SensorManager”, available at: http://developer.android.com/reference/android/hardware/SensorManager.html (accessed 8 January 2014)
〔13〕 X. Wu., V. Kumar, J.R. Quinlan, J. Ghosh, Q. Yang, H. Motoda, et al., “Top 10 algorithms in data mining”, Knowledge and Information Systems, vol. 14, pp. 1-37, 2008
〔14〕 Yijun Sun , Jian Li, “Iterative RELIEF for feature weighting“, Proceedings of the 23rd international conference on Machine learning, p.913-920, Pittsburgh, Pennsylvania, June 2006
〔15〕 K. Revett, H. Jahankhani, S. Magalhães, and H. Santos, “A survey of user authentication based on mouse dynamics,” Communications in Computer and Information Science (Global E-Security), vol. 12, pp. 210-219, 2008.