基于核主成分分析的异常轨迹检测方法

doi:10.11772/j.issn.1001-9081.2014.07.2107

计算机应用 ›› 2014, Vol. 34 ›› Issue (7): 2107-2110.DOI: 10.11772/j.issn.1001-9081.2014.07.2107

基于核主成分分析的异常轨迹检测方法

鲍苏宁,张磊,杨光

中国矿业大学计算机科学与技术学院，江苏徐州 221116

收稿日期:2014-01-13 修回日期:2014-03-02 出版日期:2014-07-01 发布日期:2014-08-01
通讯作者: 鲍苏宁
作者简介:鲍苏宁(1991-)，男，江苏泰州人，硕士研究生，主要研究方向：移动对象轨迹数据挖掘；张磊(1977-)，男，江苏沛县人，副教授，博士，主要研究方向：移动对象轨迹数据挖掘；杨光(1988-)，女，山东济宁人，硕士研究生，主要研究方向：移动对象轨迹数据挖掘。
基金资助:
中央高校基本科研业务费专项资金资助项目;教育部博士点基金资助项目;江苏省自然科学基金资助项目

Trajectory outlier detection method based on kernel principal component analysis

BAO Suning,ZHANG Lei,YANG Guang

School of Computer Science and Technology, China University of Mining and Technology, Xuzhou Jiangsu 221116, China

Received:2014-01-13 Revised:2014-03-02 Online:2014-07-01 Published:2014-08-01
Contact: BAO Suning
Supported by:
The Ph.D. Programs Foundation of Ministry of Education of China;Jiangsu postdoctoral fund;Youth Science and Technology Foundation of China University Of Mining And Technology

摘要/Abstract

摘要：

针对现有算法不能有效应用于多因素轨迹异常检测的问题，提出基于核主成分分析(KPCA)的异常轨迹检测方法。首先，为了改善轨迹特征提取的效果，采用KPCA对轨迹数据进行空间转换，将非线性空间转换到高维线性空间；其次，为了提高异常检测的准确率，采用一类支持向量机对轨迹特征数据进行无监督学习和预测；最终检测出具有异常行为的轨迹。采用大西洋飓风数据对算法进行测试，实验结果表明，该算法能够有效提取出轨迹特征，并且与同类算法相比，该算法在多因素轨迹异常检测方面具有更好的检测效果。

Abstract:

In view of the fact that the existing algorithms cannot effectively be applied to multi-factor trajectory outlier detection, this paper proposed a new method named TOD-KPCA (Trajectory Outlier Detection method based on Kernel Principal Component Analysis). Firstly, in order to enhance the effect of trajectory feature extraction, the method used KPCA to do the space transformation for trajectories and converted nonlinear space to a high dimension linear space. Furthermore, in order to improve the accuracy of outlier detection, the method used one-class Support Vector Machine (SVM) to do unsupervised learning and prediction with trajectory feature data. Finally, the method detected those trajectories with abnormal behavior. The proposed algorithm was tested on the Atlantic hurricane data. The experimental results show that the proposed algorithm can effectively extract trajectory features, and compared with the same algorithm, the proposed algorithm has better detection results in terms of multi-factor trajectory outlier detection.

中图分类号:

TP312

鲍苏宁张磊杨光. 基于核主成分分析的异常轨迹检测方法[J]. 计算机应用, 2014, 34(7): 2107-2110.

BAO Suning ZHANG Lei YANG Guang. Trajectory outlier detection method based on kernel principal component analysis[J]. Journal of Computer Applications, 2014, 34(7): 2107-2110.

参考文献

［1］International Telecommunication Union. World Telecommunication/ICT development report 2010［EB/OL］.［2013-10-12］.http://www.itu.int/ITU-D/ict/publications/wtdr_10/index.html
［2］LEE J G, HAN J, LI X. Trajectory outlier detection: A partition-and-detect framework ［C］// ICDE2008: Proceedings of the IEEE 24th International Conference on Data Engineering. Piscataway: IEEE, 2008:140-149.
［3］LIU L, QIAO S, LIU B, et al. Efficient trajectory outlier detection algorithm based on R-tree ［J］. Journal of Software, 2009, 20(9): 2426-2435.(刘良旭,乔少杰,刘宾,等.基于R-Tree的高效异常轨迹检测算法［J］.软件学报,2009,20(9):2426-2435.)
［4］XIONG L, POCZOS B, SCHNEIDER J G, et al. Hierarchical probabilistic models for group anomaly detection［C］// AISTATS2011: Proceedings of the 4th International Conference on Artificial Intelligence and Statistics. Fort Lauderdale: Microtome Publishing, 2011: 789-797.
［5］OLIVA J B. Anomaly detection and modeling of trajectories［D］. Pittsburgh: Carnegie Mellon University, 2012.
［6］CHAWLA N V, BOWYER K W, HALL L O, et al. SMOTE: Synthetic minority over-sampling technique［J］. Journal of Artificial Intelligence Research, 2002,16(1): 326-331.
［7］JAIN A, NANDAKUMAR K, ROSS A. Score normalization in multimodal biometric systems ［J］. Pattern Recognition, 2005, 38(12): 2270-2285.
［8］SCHOLKOPF B, SMOLA A, MULLER K R. Nonlinear component analysis as a kernel eigenvalue problem［J］. Neural Computation, 1998, 10(5): 1299-1319.
［9］BISHOP C M. Pattern recognition and machine learning ［M］. New York: Springer, 2006.
［10］HAN X. The convergence of the QL algorithm and QR algorithm with shifts for symmetric tridiagonal matrix［J］. Higher School Journal of Computational Mathematics, 1995,20(2):1-5. (韩旭里.对称三对角矩阵带位移的QL方法和QR方法的收敛性［J］.高等学校计算数学学报, 1995,20(2):1-5
［11］JOACHIMS T. Text categorization with support vector machines: Learning with many relevant features ［M］. Berlin: Springer, 1998.
［12］PICIARELLI C, FORESTI G L. Anomalous trajectory detection using support vector machines［C］// AVSS2007: Proceedings of Advanced Video and Signal Based Surveillance. Piscataway: IEEE, 2007:153-158.
［13］Tropical Prediction Center. Atlantic tropical storm tracking by year［EB/OL］.［2013-09-15］. http://weather.unisys.com/hurricane/atlantic/
［14］XU J, TAO X. One-class intrusion detection system based on KPCA space-similarity［J］. Journal of Computer Applications, 2009, 29(9):2460-2460.(徐晶,陶新民.基于KPCA空间相似度的一类入侵检测方法［J］.计算机应用,2009,29(9):2460-2460.)

[1]	田大江李成扬黄天波文伟平. 基于底层虚拟机的标识符混淆方法[J]. 计算机应用, 0, (): 0-0.
[2]	崔双双, 王宏志. 基于日志结构合并树的轻量级分布式索引实现方法[J]. 计算机应用, 2021, 41(3): 630-635.
[3]	樊仲欣. 基于数据流的聚类趋势分析算法[J]. 计算机应用, 2020, 40(8): 2248-2254.
[4]	刘永相李彦斌林亮江冰刘期烈谢冬菊. 基于区块链的共享汽车智能合约算法设计[J]. 计算机应用, 0, (): 0-0.
[5]	申光鹏王挺. 基于图像特征的电子印章数字水印算法设计[J]. 计算机应用, 0, (): 0-0.
[6]	谢琪, 徐旭, 程耕国, 陈和平. 基于新的森林优化算法的特征选择算法[J]. 计算机应用, 2020, 40(5): 1266-1271.
[7]	谢琪徐旭程耕国陈和平. 新的森林优化算法的特征选择算法[J]. 计算机应用, 0, (): 0-0.
[8]	樊仲欣, 王兴, 苗春生. 基于连通距离和连通强度的BIRCH改进算法[J]. 计算机应用, 2019, 39(4): 1027-1031.
[9]	郑振涛, 赵卓峰, 王桂玲, 徐垚. 面向港口停留区域识别的船舶停留轨迹提取方法[J]. 计算机应用, 2019, 39(1): 113-117.
[10]	李敏佳, 许国艳, 朱帅, 张网娟. 基于结构洞和度折扣的影响力最大化算法[J]. 计算机应用, 2018, 38(12): 3419-3424.
[11]	韩忠华毕开元司雯吕哲. 基于谱分析的密度峰值快速聚类方法算法[J]. 计算机应用, 0, (): 0-0.
[12]	李敏佳. 基于结构洞的社会网络影响力最大化算法[J]. 计算机应用, 0, (): 0-0.
[13]	李阅志, 祝园园, 钟鸣. 基于k-核过滤的社交网络影响最大化算法[J]. 计算机应用, 2018, 38(2): 464-470.
[14]	温占考, 易秀双, 田申申, 李婕, 王兴伟. 基于边界矩阵低阶近似和近邻模型的协同过滤算法[J]. 计算机应用, 2017, 37(12): 3472-3476.
[15]	常天佑, 魏强, 耿洋洋. 基于状态转换的PLC程序模型构建方法[J]. 计算机应用, 2017, 37(12): 3574-3580.

基于核主成分分析的异常轨迹检测方法

Trajectory outlier detection method based on kernel principal component analysis

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics