基于改进核可能性C均值类间极大化聚类算法An improved kernel maximum center interval possibilitic C-means clustering algorithm
林嘉炜,祁云嵩,陈晓利,凡甲甲
摘要(Abstract)：
核可能性C均值(KPCM)聚类算法只考虑类内元素之间的关系,而忽略类与类之间的关系,在对边界模糊的数据集进行聚类时会出现聚类中心距离过小甚至出现聚类中心重合的现象。针对上述问题,提出一种基于改进核可能性C均值类间极大化(KMPCM)聚类算法。该算法在核可能性C均值聚类算法上引入高维特征空间的类间极大惩罚项和调控因子λ,构造新的目标函数。这样既可以合理地拉大类中心间距离,较好地避免聚类中心距离过小甚至重合的现象,使得边界处的样本得到了较好的划分,同时也考虑类内元素的关系,保持对噪声点和野值点较好的鲁棒性。通过大量实验证明,改进算法对于边界模糊的数据集的聚类效果明显优于传统聚类算法。
关键词(KeyWords)： 核可能性C均值;边界模糊;聚类算法;类间极大惩罚项;调控因子;类内元素
基金项目(Foundation): 国家自然科学基金项目(61471182);; 2017年江苏省研究生实践创新计划(SJCX17_0607)~~
作者(Author): 林嘉炜,祁云嵩,陈晓利,凡甲甲
DOI: 10.16652/j.issn.1004-373x.2018.24.029
参考文献(References)：
- [1] LZAKIAN H,PEDRYCZ W,JAMAL I. Fuzzy clustering of time series data using dynamic time warping distance[J]. Engineering applications of artificial intelligence,2015,39:235-244.
- [2]孙如英,韩荣苍.基于FCM的模糊粗糙属性约简[J].现代电子技术,2009,32(17):194-196.SUN Ruying, HAN Rongcang. Attribute reduction approach based on fuzzy rough set and FCM[J]. Modern electronics technique,2009,32(17):194-196.
- [3] LIU Z M,LI S Z,LIN D Z,et al. Blog community discovery based on PCM clustering algorithm[J]. Journal of Xiamen University,2009,48(4):508-513.
- [4] WANG X. KFCM algorithm based on the source code mining method study[C]//Proceedings of 5th International Conference on Intelligent Systems Design and Engineering Applications.Changsha:IEEE,2014:586-588.
- [5] MA Z T,GAO J W,QIN Y,et al. Fault diagnosis of metro vehicle auxiliary inverter based on PSO-KPCM algorithm[J]. Applied mechanics&materials,2013,385:593-596.
- [6] LIU J,MOHAMMED J,CARTER J,et al. Distance-based clustering of CGH data[J]. Bioinformatics,2006,22(16):1971-1978.
- [7] PAL N R,PAL K,KELLER J M,et al. A possibilistic fuzzy Cmeans clustering algorithm[J]. IEEE transactions on fuzzy systems,2005,13(4):517-530.
- [8] ZADEH L A. Fuzzy sets[J]. Information and control,1965,8(3):338-353.
- [9] YOSHIKAWA Y,IWATA T,SAWADA H. Non-linear regression for bag-of-words data via Gaussian process latent variable set model[C/OL].[2015-02-21]. https://www.aaai.org/ocs/index.php/AAAI/AAAI15/paper/view/9796.
- [10] LIANG G E,LANG J T,TANG H,et al. Clustering high-dimensional data using PCA-Hubness[J]. Modern computer,2017(11):52-55.