Techniques of cluster algorithms in data mining

被引：195
作者：
Grabmeier, J
Rudolph, A
机构：
[1] Univ Appl Sci, D-94469 Deggendorf, Germany
[2] Univ Bundeswehr Munchen, D-85579 Neubiberg, Germany
来源：
DATA MINING AND KNOWLEDGE DISCOVERY | 2002年 / 6卷 / 04期
关键词：
data mining; cluster algorithm; Condorcet's criterion; demographic clustering;
D O I：
10.1023/A:1016308404627
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
An overview of cluster analysis techniques from a data mining point of view is given. This is done by a strict separation of the questions of various similarity and distance measures and related optimization criteria for clusterings from the methods to create and modify clusterings themselves. In addition to this general setting and overview, the second focus is used on discussions of the essential ingredients of the demographic cluster algorithm of IBM's Intelligent Miner, based Condorcet's criterion.
引用
收藏
页码：303 / 360
页数：58
相关论文
共 38 条
[21]  
MCLACHLAN GJ, MIXTURE MODELS
[22]  
MESSATFA H, 1997, FUTURE GEN COMPUTER, V13, P149
[23]   Clustering techniques [J].
Michaud, P .
FUTURE GENERATION COMPUTER SYSTEMS, 1997, 13 (2-3) :135-147
[24]  
Michaud P., 1987, APPL STOCH MODEL DAT, V3, P173, DOI [10.1002/asm.3150030305, DOI 10.1002/ASM.3150030305]
[25]  
MICHAUD P, 1985, F052 IBM CTR SCI
[26]  
MICHAUD P, 1995, MAP010 IBM ECAM
[27]  
MICHAUD P, 1982, F051 IBM CTR SCI
[28]  
MICHAUD P, 1987, F084 IBM CTR SCI
[29]  
Rao C. R., 1965, LINEAR STAT INFERENC
[30]  
RENYI A, 1962, WAHRSCHEINLICHKEITST
1234