EFFECT OF DIFFERENT PREPROCESSING METHODS FOR PRINCIPAL COMPONENT ANALYSIS APPLIED TO THE COMPOSITION OF MIXTURES - DETECTION OF IMPURITIES IN HPLC-DAD

被引：22

作者：

SANCHEZ, FC

LEWI, PJ

MASSART, DL

机构：

[1] VRIJE UNIV BRUSSELS,CHEMOAC,LAARBEEKLAAN 103,B-1090 BRUSSELS,BELGIUM

[2] JANSSEN RES FDN,DEPT INFORMAT SCI,B-2340 BEERSE,BELGIUM

来源：

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS | 1994年 / 25卷 / 02期

关键词：

D O I：

10.1016/0169-7439(94)85040-2

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

Principal component analysis is applied to the detection of impure and pure zones in the chromatograms and the spectra obtained by high-performance liquid chromatography coupled with diode-array detection. Transformation of the raw data is one of the most important steps in multivariate analysis techniques. Different preprocessing methods: no transformation, column centering, column standardization, selective normalization, log column centering, log row centering, log double centering, and double closure have been applied to the data. The biplots obtained for the two first principal components are discussed and particular attention is paid to the results obtained by the logarithmic transformation methods.

引用

页码：157 / 177

页数：21

共 23 条

[1] CALCULATION OF EQUILIBRIUM-CONSTANTS FROM MULTIWAVELENGTH SPECTROSCOPIC DATA .4. MODEL-FREE LEAST-SQUARES REFINEMENT BY USE OF EVOLVING FACTOR-ANALYSIS [J].