PSEUDOINVERSE IN CLUSTERING PROBLEMS

被引：2

作者：

Kirichenko, N. F. ^{[1
]}

Donchenko, V. S. ^{[2
]}

机构：

[1] Natl Acad Sci Ukraine, VM Glushkov Inst Cybernet, Kiev, Ukraine

[2] Taras Shevchenko Natl Univ, Kiev, Ukraine

来源：

CYBERNETICS AND SYSTEMS ANALYSIS | 2007年 / 43卷 / 04期

关键词：

Moore-Penrose pseudoinverse; singular value decomposition (SVD); orthogonal projectors; pseudoinverse of perturbations of matrices; clustering; hyperplane clustering;

D O I：

10.1007/s10559-007-0078-y

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

The clustering of vector observations of hyperplanes is studied. Different cases of correspondence distances are proposed and investigated, including the algebraic Jack Knife one. The efficiency, constructivity, and explicit form of formulas are provided by using the pseudoinverse technique including the pseudoinverse-perturbation theory. Results important for the application of pseudoinverse and related operators are presented.

引用

页码：527 / 541

页数：15

共 11 条

[1]

ALBERT A, 1977, REGRESSION PSEUDOINV

[2]

Donchenko V. S., 2006, P 12 INT KNOWL DIAL, P63

[3]

Efron B., 1988, NONCONVENTIONAL METH

[4]

Haykin S., 1999, NEURAL NETW COMPR FD, V2, P443

[5]

Kirichenko M. F., 2005, ZH OBCHYSL PRYKL MAT, P63

[6] Application of Pseudoinverse and Projective Matrices to Studying Control, Observation, and Identification Problems [J].

N. F. Kirichenko ;

N. P. Lepekha .

Cybernetics and Systems Analysis, 2002, 38 (4) :568-585

[7] NONLINEAR RECURSIVE REGRESSION TRANSFORMERS: DYNAMIC SYSTEMS AND OPTIMIZATION [J].

Kirichenko, N. F. ;

Donchenko, V. S. ;

Serbaev, D. P. .

CYBERNETICS AND SYSTEMS ANALYSIS, 2005, 41 (03) :364-373

[8] Analytical representation of perturbations of pseudoinverse matrices [J].

Kirichenko, NF .

CYBERNETICS AND SYSTEMS ANALYSIS, 1997, 33 (02) :230-238

[9]

Moore EH., 1920, B AM MATH SOC, V26, P394, DOI [10.1090/S0002-9904-1920-03322-7, DOI 10.1090/S0002-9904-1920-03322-7]

[10]

Penrose R., 1955, P CAMB PHIL SOC, V51, P406, DOI [10.1017/S0305004100030401, DOI 10.1017/S0305004100030401]

← 1 2 →