Support vector clustering of time series data with alignment kernels

被引:8
作者
Boecking, Benedikt [2 ]
Chalup, Stephan K. [1 ]
Seese, Detlef [2 ]
Wong, Aaron S. W. [1 ]
机构
[1] Univ Newcastle, Sch Elect Engn & Comp Sci, Callaghan, NSW 2308, Australia
[2] Karlsruhe Inst Technol, Inst Appl Informat & Formal Descript Methods AIFB, D-76128 Karlsruhe, Germany
关键词
Support vector clustering; Time series; Alignment kernel; Clustering; LABELING METHOD;
D O I
10.1016/j.patrec.2014.03.015
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Time series clustering is an important data mining topic and a challenging task due to the sequences' potentially very complex structures. In the present study we experimentally investigate the combination of support vector clustering with a triangular alignment kernel by evaluating it on an artificial time series benchmark dataset. The experiments lead to meaningful segmentations of the data, thereby providing an example that clustering time series with specific kernels is possible without pre-processing of the data. We compare our approach and the results and learn that the clustering quality is competitive when compared to other approaches. (C) 2014 Elsevier B.V. All rights reserved.
引用
收藏
页码:129 / 135
页数:7
相关论文
共 30 条
[1]  
Alcock R. J., 1999, 7 HELLENIC C INFORMA, P27
[2]  
[Anonymous], 2008, FINDING GROUPS DATA
[3]   Support vector clustering [J].
Ben-Hur, A ;
Horn, D ;
Siegelmann, HT ;
Vapnik, V .
JOURNAL OF MACHINE LEARNING RESEARCH, 2002, 2 (02) :125-137
[4]  
Camastra F., 2008, THEORY APPL ADV INFO
[5]  
Chis M, 2009, STUD COMPUT INTELL, V206, P193
[6]  
Cuturi M, 2007, INT CONF ACOUST SPEE, P413
[7]  
Cuturi Marco, 2011, P 28 INT C MACH LEAR, P929
[8]   A survey of kernel and spectral methods for clustering [J].
Filippone, Maurizio ;
Camastra, Francesco ;
Masulli, Francesco ;
Rovetta, Stefano .
PATTERN RECOGNITION, 2008, 41 (01) :176-190
[9]  
Frank A., 2010, UCI machine learning repository, V213
[10]  
Gavrilov M., 2000, Proceedings. KDD-2000. Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, P487, DOI 10.1145/347090.347189