隨著網路快速發展後，網路安全已經開始是一個很重要的議題之一，許多我們在電腦上的行為都需要一個驗證的步驟，去做個授權存取。我們研究的目標就是要
認證這個人是不是帳號的擁有者。根據輸入使用者軌跡，我們提出一個普遍的方式去驗證使用者，我們的方式不要求使用者做額外的工作，它也可能提防其它未授權使用者或惡意程式拷貝或模擬軌跡。

我們的研究著重在帳號使用者產生的軌跡中，找出隱藏的樣式，即使使用者的帳號或密碼被竊取，我們仍然能透過分析行為軌跡去決定使用者是帳號擁有者還是駭客。我們使用馬可夫鏈模型去描述軌跡，用高斯分佈去模擬兩點之間的變化。為了要去分辨兩個軌跡，我們提出一個不相似度測量，於是我們使用不相似度測量的結果，透過流線型學習去找到一對軌跡之間的關係，根據兩個軌跡不相似度，我們可以採用有效的分類或是分群方法去偵測一個未授權的存取。

The network security has become one of the important issues after
Internet expands vigorously. Many of our activities on computer
need a \textit{verification} step for authorized access. The
purpose of our study is to verify whether the person is the true
account owner. We propose a general approach for user verification
based on user trajectory inputs. The approach is labor-free for
users and is likely to avoid the possible copy or simulation from
other non-authorized users or even automatic programs like bots.

Our study focuses on finding the hidden pattern on the
trajectories produced by account users. We hope that we analyse
the behavior patterns for deciding a user which is the owner or
the cracker, although users' account and password are stolen by
hackers. We employ a Markov Chain model with Gaussian distribution
in transitions to describe the behavior on the trajectory. To
distinguish between two trajectories, we propose a novel
\textit{dissimilarity measure} combined with a manifold learnt
tuning for catching the pairwise relationships. Based on the
pairwise dissimilarities, we can plug-in any effective
classification methods or clustering methods for the detection of
unauthorized access.

[1] C. M. Bishop. Pattern Recognition and Machine Learning. Springer,
2006.
[2] L. Bovino, Sebastiano Impedovo, Giuseppe Pirlo, and L. Sarcinella.
Multi-expert veri‾cation of hand-written signatures. In ICDAR,
pages 932{936, 2003.
[3] Christopher J. C. Burges. A tutorial on support vector machines for
pattern recognition. Data Min. Knowl. Discov., 2(2):121{167, 1998.
[4] Kuan-Ta Chen, Hsing-Kuo Kenneth Pao, and Hong-Chung Chang.
Game bot identi‾cation based on manifold learning. In Proceedings
of ACM NetGames 2008, 2008.
[5] Thomas M. Cover and Joy A. Thomas. Elements of Information
Theory (2nd Ed.). Wiley-Interscience, July 2006.
[6] V. de Silva and J. B. Tenenbaum. Global versus local methods in
nonlinear dimensionality reduction. Advances in Neural Information
Processing Systems, pages 705{712, 2003.
[7] Anil K. Jain, Friederike D. Griess, and Scott D. Connell. On-line
signature veri‾cation,. Pattern Recognition, 35(12):2963{2972, 2002.
[8] Eamonn Keogh, Stefano Lonardi, and Chotirat Ann Ratanama-
hatana. Towards parameter-free data mining. In KDD '04: Proc.
of the 10th ACM SIGKDD inter. conf. on Knowledge discovery and
data mining, pages 206{215, 2004.
[9] M. H. C. Law and A. K. Jain. Incremental nonlinear dimensionality
reduction by manifold learning. IEEE Trans. on Pattern Analysis
and Machine Intelligence, 28(3):377{391, 2006.
[10] Jae-Gil Lee, Jiawei Han, Xiaolei Li, and Hector Gonzalez.
raClass: trajectory classi‾cation using hierarchical region-based and
trajectory-based clustering. PVLDB, 1(1):1081{1094, 2008.
[11] Yuh-Jye Lee and O. L. Mangasarian. Ssvm: A smooth support vector
machine for classi‾cation. Comput. Optim. Appl., 20(1):5{22, 2001.
[12] M. Li, J. H. Badger, X. Chen, S. Kwong, P. Kearney, and H. Zhang.
An information-based sequence distance and its application to whole
mitochondrial genome phylogeny. Bioinformatics, 17(2):149{154,
2001.
[13] M. Li and P. Vit¶anyi. An Introduction to Kolmogorov Complexity and
Its Applications (2nd Ed.). Springer, New York, 1997.
[14] Jessica Lin, Eamonn J. Keogh, Stefano Lonardi, and Bill Yuan chi
Chiu. A symbolic representation of time series, with implications for
streaming algorithms. pages 2{11, 2003.
[15] Heikki Mannila, Hannu Toivonen, and A. Inkeri Verkamo. Discover-
ing frequent episodes in sequences. In KDD, pages 210{215, 1995.
[16] Mario E. Munich and Pietro Perona. Visual identi‾cation by signa-
ture tracking. IEEE Trans. Pattern Anal. Mach. Intell., 25(2):200{
217, 2003.
[17] Hsing-Kuo Pao and John Case. Computing entropy for ortholog de-
tection. In International Conference on Computational Intelligence,
pages 89{92, 2004.
[18] Yu Qiao, Jianzhuang Liu, and Xiaoou Tang. O²ine signature veri‾-
cation using online handwriting registration. In CVPR, 2007.
[19] J. Richiardi and A. Drygajlo. Gaussian mixture models for on-line
signature veri‾cation. In Intl Multimedia Conf., Proc. 2003 ACM
SIGMM workshop on Biometrics methods and applications, pages
115{122, Berkeley, USA, Nov. 2003.
[20] B. SchÄolkopf and A. Smola. Learning with Kernels Support Vector
Machines, Regularization, Optimization and Beyond. MIT Press,
Cambridge, MA, USA, 2002.
[21] Claudio De Stefano, Ciro D'Elia, Alessandra Scotto di Freca, and An-
gelo Marcelli. Classi‾er combination by bayesian networks for hand-
writing recognition. IJPRAI, 23(5):887{905, 2009.
[22] Vladimir N. Vapnik. The Nature of Statistical Learning Theory.
Springer, 2 edition, November 1999.
[23] Luis von Ahn, Manuel Blum, Nicholas J. Hopper, and John Langford.
Captcha: Using hard ai problems for security. In EUROCRYPT,
pages 294{311, 2003.