基于簇特征的增量聚类算法
|
摘要
针对传统聚类算法无法处理大规模数据的特点,结合增量算法和簇特征的思想,在初始聚类阶段,采用基于距离的K-means聚类算法获取相应簇的特征。根据簇特征,并结合K最近邻(KNN)的思想处理增量,提出了基于簇特征的增量聚类算法。提出的方法已经在加州大学尔湾分校(UCI)机器学习库中提供的真实数据集的帮助下得到验证。实验结果表明:提出的增量聚类方法的聚类精度较普通K-means算法和原始增量K-means算法有明显提高。
Aiming at the feature that traditional clustering algorithm can not deal with large-scale data,an incremental clustering algorithm based on cluster feature( ICABCF) is presented combined with incremental algorithm and cluster feature idea,in the stage of initial clustering,with the help of K-means clustering algorithm based on distance,cluster feature is obtained. According to cluster feature and K-nearest neighbor( KNN)algorithm is used to process increment. The proposed approach is validated with the help of real datasets presented in the UCI machine learning repository. The experimental results demonstrate that clustering precision of the proposed incremental clustering approach is improved significantly,compared with ordinary K-means and original incremental K-means algorithms.
Aiming at the feature that traditional clustering algorithm can not deal with large-scale data,an incremental clustering algorithm based on cluster feature( ICABCF) is presented combined with incremental algorithm and cluster feature idea,in the stage of initial clustering,with the help of K-means clustering algorithm based on distance,cluster feature is obtained. According to cluster feature and K-nearest neighbor( KNN)algorithm is used to process increment. The proposed approach is validated with the help of real datasets presented in the UCI machine learning repository. The experimental results demonstrate that clustering precision of the proposed incremental clustering approach is improved significantly,compared with ordinary K-means and original incremental K-means algorithms.
引文
[1] Fisher D. Knowledge acquisition via incremental conceptual clustering[J]. Machine Learning,1987,2:139-172.
[2] Can F. Incremental clustering for dynamic information processing[J].ACM Transactions on Information Systems,1993,11(2):143-164.
[3] Contents T O. Very large-scale incremental clustering[J]. Journal of the American Oil Chemists Society,1952,29(10):4-22.
[4] Langford T,Giraud-Carrier C G,Magee J. Detection of infectious outbreaks in hospitals through incremental clustering[C]∥Conference on AI in Medicine in Europe:Artificial Intelligence Medicine,Springer-Verlag,2001:30-39.
[5] Lin J,Vlachos M,Keogh E,et al. Advances in database technology[M]. Berlin Heidelberg:Springer,2004:106-122.
[6] Simovici D,Singla N,Kuperberg M. Metric incremental clustering of nominal data[C]∥IEEE International Conference on Data Mining,IEEE,2004:523-526.
[7] Zhang T,Ramakrishnan R,Livny M. BIRCH:An efficient data clustering method for very large databases[C]∥ACM SIGMOD International Conference on Management of Data,ACM,1996:103-114.
[8]张乙竹,周礼争,唐瑞,等.基于K-means聚类点密度的WSNs加权质心定位算法[J].传感器与微系统,2015,34(7):125-127.
[9]龙慧,樊晓平,刘少强,等.无线传感器网络动态最近邻协作目标跟踪算法[J].传感器与微系统,2012,31(7):135-139.
[10]张晓峰,王丽珍,陆叶.一种基于属性加权的不确定K-means聚类算法[J].计算机研究与发展,2009,46(s2):879-883.
[2] Can F. Incremental clustering for dynamic information processing[J].ACM Transactions on Information Systems,1993,11(2):143-164.
[3] Contents T O. Very large-scale incremental clustering[J]. Journal of the American Oil Chemists Society,1952,29(10):4-22.
[4] Langford T,Giraud-Carrier C G,Magee J. Detection of infectious outbreaks in hospitals through incremental clustering[C]∥Conference on AI in Medicine in Europe:Artificial Intelligence Medicine,Springer-Verlag,2001:30-39.
[5] Lin J,Vlachos M,Keogh E,et al. Advances in database technology[M]. Berlin Heidelberg:Springer,2004:106-122.
[6] Simovici D,Singla N,Kuperberg M. Metric incremental clustering of nominal data[C]∥IEEE International Conference on Data Mining,IEEE,2004:523-526.
[7] Zhang T,Ramakrishnan R,Livny M. BIRCH:An efficient data clustering method for very large databases[C]∥ACM SIGMOD International Conference on Management of Data,ACM,1996:103-114.
[8]张乙竹,周礼争,唐瑞,等.基于K-means聚类点密度的WSNs加权质心定位算法[J].传感器与微系统,2015,34(7):125-127.
[9]龙慧,樊晓平,刘少强,等.无线传感器网络动态最近邻协作目标跟踪算法[J].传感器与微系统,2012,31(7):135-139.
[10]张晓峰,王丽珍,陆叶.一种基于属性加权的不确定K-means聚类算法[J].计算机研究与发展,2009,46(s2):879-883.