Analysis of trace elements in South African clinkers using latent variable model and clustering

J. Abonyi, Ferenc D. Tamás, Sanja Potgieter, Herman Potgieter

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The trace element content of clinkers (and possibly of cements) can be used to identify the manufacturing factory. The Mg, Sr, Ba, Mn, Ti, Zr, Zn and V content of clinkers give detailed information for the determination of the origin of clinkers produced in different factories. However, for the analysis of such complex data there is a need for algorithmic tools for the visualization and clustering of the samples. This paper proposes a new approach for this purpose. The analytical data are transformed into a two-dimensional latent space by factor analysis (probabilistic principal component analysis) and dendograms are constructed for cluster formation. The classification of South African clinkers is used as an illustrative example for the approach.

Original languageEnglish
Pages (from-to)15-20
Number of pages6
JournalSouth African Journal of Chemistry
Volume56
Publication statusPublished - May 9 2003

Fingerprint

Trace Elements
Industrial plants
Factor analysis
Principal component analysis
Cements
Visualization

Keywords

  • Clinker
  • Clustering
  • Dendogram
  • Factor analysis
  • Principal component analysis
  • Trace elements

ASJC Scopus subject areas

  • Chemistry(all)

Cite this

Analysis of trace elements in South African clinkers using latent variable model and clustering. / Abonyi, J.; Tamás, Ferenc D.; Potgieter, Sanja; Potgieter, Herman.

In: South African Journal of Chemistry, Vol. 56, 09.05.2003, p. 15-20.

Research output: Contribution to journalArticle

Abonyi, J. ; Tamás, Ferenc D. ; Potgieter, Sanja ; Potgieter, Herman. / Analysis of trace elements in South African clinkers using latent variable model and clustering. In: South African Journal of Chemistry. 2003 ; Vol. 56. pp. 15-20.
@article{20d38abdd22d45bea433571dd04f3d4d,
title = "Analysis of trace elements in South African clinkers using latent variable model and clustering",
abstract = "The trace element content of clinkers (and possibly of cements) can be used to identify the manufacturing factory. The Mg, Sr, Ba, Mn, Ti, Zr, Zn and V content of clinkers give detailed information for the determination of the origin of clinkers produced in different factories. However, for the analysis of such complex data there is a need for algorithmic tools for the visualization and clustering of the samples. This paper proposes a new approach for this purpose. The analytical data are transformed into a two-dimensional latent space by factor analysis (probabilistic principal component analysis) and dendograms are constructed for cluster formation. The classification of South African clinkers is used as an illustrative example for the approach.",
keywords = "Clinker, Clustering, Dendogram, Factor analysis, Principal component analysis, Trace elements",
author = "J. Abonyi and Tam{\'a}s, {Ferenc D.} and Sanja Potgieter and Herman Potgieter",
year = "2003",
month = "5",
day = "9",
language = "English",
volume = "56",
pages = "15--20",
journal = "South African Journal of Chemistry",
issn = "0379-4350",
publisher = "Bureau for Scientific Publications",

}

TY - JOUR

T1 - Analysis of trace elements in South African clinkers using latent variable model and clustering

AU - Abonyi, J.

AU - Tamás, Ferenc D.

AU - Potgieter, Sanja

AU - Potgieter, Herman

PY - 2003/5/9

Y1 - 2003/5/9

N2 - The trace element content of clinkers (and possibly of cements) can be used to identify the manufacturing factory. The Mg, Sr, Ba, Mn, Ti, Zr, Zn and V content of clinkers give detailed information for the determination of the origin of clinkers produced in different factories. However, for the analysis of such complex data there is a need for algorithmic tools for the visualization and clustering of the samples. This paper proposes a new approach for this purpose. The analytical data are transformed into a two-dimensional latent space by factor analysis (probabilistic principal component analysis) and dendograms are constructed for cluster formation. The classification of South African clinkers is used as an illustrative example for the approach.

AB - The trace element content of clinkers (and possibly of cements) can be used to identify the manufacturing factory. The Mg, Sr, Ba, Mn, Ti, Zr, Zn and V content of clinkers give detailed information for the determination of the origin of clinkers produced in different factories. However, for the analysis of such complex data there is a need for algorithmic tools for the visualization and clustering of the samples. This paper proposes a new approach for this purpose. The analytical data are transformed into a two-dimensional latent space by factor analysis (probabilistic principal component analysis) and dendograms are constructed for cluster formation. The classification of South African clinkers is used as an illustrative example for the approach.

KW - Clinker

KW - Clustering

KW - Dendogram

KW - Factor analysis

KW - Principal component analysis

KW - Trace elements

UR - http://www.scopus.com/inward/record.url?scp=4243190121&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=4243190121&partnerID=8YFLogxK

M3 - Article

VL - 56

SP - 15

EP - 20

JO - South African Journal of Chemistry

JF - South African Journal of Chemistry

SN - 0379-4350

ER -