Early transition detection - A dynamic extension to common classification methods

Authors
Marth, M Maier, D Honerkamp, J Rupprecht, M Goschnick, J
Citation
M. Marth et al., Early transition detection - A dynamic extension to common classification methods, CHEM INTELL, 43(1-2), 1998, pp. 123-133
Citations number
18
Categorie Soggetti
Spectroscopy /Instrumentation/Analytical Sciences
Journal title
CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS
ISSN journal
01697439 → ACNP
Volume
43
Issue
1-2
Year of publication
1998
Pages
123 - 133
Database
ISI
SICI code
0169-7439(19980928)43:1-2<123:ETD-AD>2.0.ZU;2-K
Abstract
An extension for classification methods in order to process time-dependent data is introduced. It is based on the detection of transitions from one st eady state to another one by examination of the time derivatives of classif ication vectors. The method is called Early Transition Detection (ETD). It is shown that it can be used in conjunction with a number of common classif ication methods like SIMCA or Artificial Neural Nets and it is successfully tested on simulated and on real data. (C) 1998 Elsevier Science B.V. All r ights reserved.