基于复合形的K-means优化聚类算法研究
|
摘要
为了克服K-means算法受初始点影响大、结果稳定性差的不足,提出了一种新的K-means优化聚类算法.介绍了复合形法的基本原理并将其做了一定修改以适用于K-means优化聚类,推导了一系列用于计算的公式,给出了具体的实现步骤与方法.通过算例说明,与其他几种方法相比,该方法结果稳定,计算效率较高,有着很好的推广应用前景.
A novel K-means algorithm is presented to overcome the weakness of the result stabilization and the effect of the initial point.First,the main principle of the complex method is introduced and an improved one is used to K-means optimal clustering,a series of formulation is educed in this paper and the realization process of the algorithm is given later.Compared with other algorithms,this method has a stable result,an efficient calculation,so it has very good application and dissemination prospect.
A novel K-means algorithm is presented to overcome the weakness of the result stabilization and the effect of the initial point.First,the main principle of the complex method is introduced and an improved one is used to K-means optimal clustering,a series of formulation is educed in this paper and the realization process of the algorithm is given later.Compared with other algorithms,this method has a stable result,an efficient calculation,so it has very good application and dissemination prospect.
引文
[1]Qu Shouning,Wang Qin,Zou Yan,et al.Intelligent question answering system based on data mining[J].Journal of Zheng-zhou University:Natural Science Edition,2007,39(2):50-54.
[2]华斌,陈玲.基于聚类的科研经费决策与分析模型[J].郑州大学学报:理学版,2007,39(3):55-58.
[3]胡玉锁,陈宗海.基于混合遗传算法的聚类分析[J].模式识别与人工智能,2001,14(3):352-356.
[4]MacQueen J.Some methods for classification and analysis of multi-variate observations[C]∥Proceedings of the 15thBerkeley Symposium on Mathematical Statistics and Probability.Berkeley:University of California Press,1967:281-297.
[5]Bandyopadhyay S,Maulik U.An evolutionary technique based on K-means algorithm for optional clustering inRN[J].In-formation Sciences,2002,146(1/2/3/4):221-237.
[6]欧海龙,李向真.多层复形法在结构优化中的程序设计[J].世界地震工程,2000,16(4):113-117.
[7]苏金明,阮沈勇,王永利.MATLAB工程数学[M].北京:电子工业出版社,2005.
[8]朱金钧,高凯,周万珍.遗传算法在数据挖掘中的应用[J].计算机工程与应用,2003,39(17):203-206.
[9]屈建平,罗文坚,王煦法.基于K-均值聚类的改进非选择算法研究[J].计算机工程与应用,2005,41(28):29-32.
[10]Krishna K,Murty M N.Genetic K-means algorithm[J].IEEE Transactions on Systems,Man and Cybernetics-Part B:Cy-bernetics,1998,29(5):433-439.
[11]Selim S Z,Ismail M A.K-means-type algorithms:a generalized convergence theorem and characterization of local opti-ma1ity[J].IEEE Trans Pattern Analysis and Machine Intelligence,1984(1):81-87.
[2]华斌,陈玲.基于聚类的科研经费决策与分析模型[J].郑州大学学报:理学版,2007,39(3):55-58.
[3]胡玉锁,陈宗海.基于混合遗传算法的聚类分析[J].模式识别与人工智能,2001,14(3):352-356.
[4]MacQueen J.Some methods for classification and analysis of multi-variate observations[C]∥Proceedings of the 15thBerkeley Symposium on Mathematical Statistics and Probability.Berkeley:University of California Press,1967:281-297.
[5]Bandyopadhyay S,Maulik U.An evolutionary technique based on K-means algorithm for optional clustering inRN[J].In-formation Sciences,2002,146(1/2/3/4):221-237.
[6]欧海龙,李向真.多层复形法在结构优化中的程序设计[J].世界地震工程,2000,16(4):113-117.
[7]苏金明,阮沈勇,王永利.MATLAB工程数学[M].北京:电子工业出版社,2005.
[8]朱金钧,高凯,周万珍.遗传算法在数据挖掘中的应用[J].计算机工程与应用,2003,39(17):203-206.
[9]屈建平,罗文坚,王煦法.基于K-均值聚类的改进非选择算法研究[J].计算机工程与应用,2005,41(28):29-32.
[10]Krishna K,Murty M N.Genetic K-means algorithm[J].IEEE Transactions on Systems,Man and Cybernetics-Part B:Cy-bernetics,1998,29(5):433-439.
[11]Selim S Z,Ismail M A.K-means-type algorithms:a generalized convergence theorem and characterization of local opti-ma1ity[J].IEEE Trans Pattern Analysis and Machine Intelligence,1984(1):81-87.
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心