A Support Vector and K-Means Based Hybrid Intelligent Data Clustering Algorithm

Liang SUN; Shinichi YOSHIDA; Yanchun LIANG

doi:10.1587/transinf.E94.D.2234

A Support Vector and K-Means Based Hybrid Intelligent Data Clustering Algorithm

Liang SUN, Shinichi YOSHIDA, Yanchun LIANG

Full Text Views

0

Cite this

Summary :

Support vector clustering (SVC), a recently developed unsupervised learning algorithm, has been successfully applied to solving many real-life data clustering problems. However, its effectiveness and advantages deteriorate when it is applied to solving complex real-world problems, e.g., those with large proportion of noise data points and with connecting clusters. This paper proposes a support vector and K-Means based hybrid algorithm to improve the performance of SVC. A new SVC training method is developed based on analysis of a Gaussian kernel radius function. An empirical study is conducted to guide better selection of the standard deviation of the Gaussian kernel. In the proposed algorithm, firstly, the outliers which increase problem complexity are identified and removed by training a global SVC. The refined data set is then clustered by a kernel-based K-Means algorithm. Finally, several local SVCs are trained for the clusters and then each removed data point is labeled according to the distance from it to the local SVCs. Since it exploits the advantages of both SVC and K-Means, the proposed algorithm is capable of clustering compact and arbitrary organized data sets and of increasing robustness to outliers and connecting clusters. Experiments are conducted on 2-D data sets generated by mixture models and benchmark data sets taken from the UCI machine learning repository. The cluster error rate is lower than 3.0% for all the selected data sets. The results demonstrate that the proposed algorithm compared favorably with existing SVC algorithms.

Publication: IEICE TRANSACTIONS on Information Vol.E94-D No.11 pp.2234-2243

Publication Date: 2011/11/01

Publicized

Online ISSN: 1745-1361

DOI: 10.1587/transinf.E94.D.2234

Type of Manuscript: PAPER

Category: Artificial Intelligence, Data Mining

Cite this

Copy

Liang SUN, Shinichi YOSHIDA, Yanchun LIANG, "A Support Vector and K-Means Based Hybrid Intelligent Data Clustering Algorithm" in IEICE TRANSACTIONS on Information, vol. E94-D, no. 11, pp. 2234-2243, November 2011, doi: 10.1587/transinf.E94.D.2234.
Abstract: Support vector clustering (SVC), a recently developed unsupervised learning algorithm, has been successfully applied to solving many real-life data clustering problems. However, its effectiveness and advantages deteriorate when it is applied to solving complex real-world problems, e.g., those with large proportion of noise data points and with connecting clusters. This paper proposes a support vector and K-Means based hybrid algorithm to improve the performance of SVC. A new SVC training method is developed based on analysis of a Gaussian kernel radius function. An empirical study is conducted to guide better selection of the standard deviation of the Gaussian kernel. In the proposed algorithm, firstly, the outliers which increase problem complexity are identified and removed by training a global SVC. The refined data set is then clustered by a kernel-based K-Means algorithm. Finally, several local SVCs are trained for the clusters and then each removed data point is labeled according to the distance from it to the local SVCs. Since it exploits the advantages of both SVC and K-Means, the proposed algorithm is capable of clustering compact and arbitrary organized data sets and of increasing robustness to outliers and connecting clusters. Experiments are conducted on 2-D data sets generated by mixture models and benchmark data sets taken from the UCI machine learning repository. The cluster error rate is lower than 3.0% for all the selected data sets. The results demonstrate that the proposed algorithm compared favorably with existing SVC algorithms.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E94.D.2234/_p

Copy

@ARTICLE{e94-d_11_2234,
author={Liang SUN, Shinichi YOSHIDA, Yanchun LIANG, },
journal={IEICE TRANSACTIONS on Information},
title={A Support Vector and K-Means Based Hybrid Intelligent Data Clustering Algorithm},
year={2011},
volume={E94-D},
number={11},
pages={2234-2243},
abstract={Support vector clustering (SVC), a recently developed unsupervised learning algorithm, has been successfully applied to solving many real-life data clustering problems. However, its effectiveness and advantages deteriorate when it is applied to solving complex real-world problems, e.g., those with large proportion of noise data points and with connecting clusters. This paper proposes a support vector and K-Means based hybrid algorithm to improve the performance of SVC. A new SVC training method is developed based on analysis of a Gaussian kernel radius function. An empirical study is conducted to guide better selection of the standard deviation of the Gaussian kernel. In the proposed algorithm, firstly, the outliers which increase problem complexity are identified and removed by training a global SVC. The refined data set is then clustered by a kernel-based K-Means algorithm. Finally, several local SVCs are trained for the clusters and then each removed data point is labeled according to the distance from it to the local SVCs. Since it exploits the advantages of both SVC and K-Means, the proposed algorithm is capable of clustering compact and arbitrary organized data sets and of increasing robustness to outliers and connecting clusters. Experiments are conducted on 2-D data sets generated by mixture models and benchmark data sets taken from the UCI machine learning repository. The cluster error rate is lower than 3.0% for all the selected data sets. The results demonstrate that the proposed algorithm compared favorably with existing SVC algorithms.},
keywords={},
doi={10.1587/transinf.E94.D.2234},
ISSN={1745-1361},
month={November},}

Copy

TY - JOUR
TI - A Support Vector and K-Means Based Hybrid Intelligent Data Clustering Algorithm
T2 - IEICE TRANSACTIONS on Information
SP - 2234
EP - 2243
AU - Liang SUN
AU - Shinichi YOSHIDA
AU - Yanchun LIANG
PY - 2011
DO - 10.1587/transinf.E94.D.2234
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E94-D
IS - 11
JA - IEICE TRANSACTIONS on Information
Y1 - November 2011
AB - Support vector clustering (SVC), a recently developed unsupervised learning algorithm, has been successfully applied to solving many real-life data clustering problems. However, its effectiveness and advantages deteriorate when it is applied to solving complex real-world problems, e.g., those with large proportion of noise data points and with connecting clusters. This paper proposes a support vector and K-Means based hybrid algorithm to improve the performance of SVC. A new SVC training method is developed based on analysis of a Gaussian kernel radius function. An empirical study is conducted to guide better selection of the standard deviation of the Gaussian kernel. In the proposed algorithm, firstly, the outliers which increase problem complexity are identified and removed by training a global SVC. The refined data set is then clustered by a kernel-based K-Means algorithm. Finally, several local SVCs are trained for the clusters and then each removed data point is labeled according to the distance from it to the local SVCs. Since it exploits the advantages of both SVC and K-Means, the proposed algorithm is capable of clustering compact and arbitrary organized data sets and of increasing robustness to outliers and connecting clusters. Experiments are conducted on 2-D data sets generated by mixture models and benchmark data sets taken from the UCI machine learning repository. The cluster error rate is lower than 3.0% for all the selected data sets. The results demonstrate that the proposed algorithm compared favorably with existing SVC algorithms.
ER -