Molecular dynamics-like data clustering approach

被引：13

作者：

Li Junlin ^{[1
]}

Fu Hongguang ^{[1
]}

机构：

[1] Univ Elect Sci & Technol China, Sch Comp Sci & Engn, Chengdu 610054, Peoples R China

来源：

PATTERN RECOGNITION | 2011年 / 44卷 / 08期

关键词：

Molecular dynamics; Dynamics clustering; Data mining; Data clustering; K-MEANS;

D O I：

10.1016/j.patcog.2011.01.008

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Based on the molecular kinetic theory, a molecular dynamics-like data clustering approach is proposed in this paper. Clusters are extracted after data points fuse in the iterating space by the dynamical mechanism that is similar to the interacting mechanism between molecules through molecular forces. This approach is to find possible natural clusters without pre-specifying the number of clusters. Compared with 3 other clustering methods (trimmed k-means, JP algorithm and another gravitational model based method), this approach found clusters better than the other 3 methods in the experiments. (C) 2011 Elsevier Ltd. All rights reserved.

引用

页码：1721 / 1737

页数：17

共 26 条

[1] [Anonymous], 1988, ALGORITHMS CLUSTERIN
[2] Binder K., 1995, Monte Carlo and Molecular Dynamics Simulations in Polymer Science
[3] Chen Zhenglong, 2007, THEORIES PRACTICES M
[4] Cuesta-Albertos JA, 1997, ANN STAT, V25, P553
[5] CHARACTERIZATION AND DETECTION OF NOISE IN CLUSTERING
DAVE, RN
[J]. PATTERN RECOGNITION LETTERS, 1991, 12 (11) : 657 - 664
[6] Endo Y, 2005, LECT NOTES ARTIF INT, V3558, P183
[7] Robustness properties of k means and trimmed k means
García-Escudero, LA
Gordaliza, A
[J]. JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1999, 94 (447) : 956 - 969
[8] Gomez J, 2003, SIAM PROC S, P83
[9] Hinneburg A., 1998, Proceedings Fourth International Conference on Knowledge Discovery and Data Mining, P58
[10] [蒋盛益 Jiang Shengyi], 2005, [计算机应用, Computer Applications], V25, P286

← 1 2 3 →