Abstract Correspondence cluster analysis, developed in this paper, regroups the main advantages of correspondence analysis and cluster analysis. The output of the method, correspondence cluster dendrograms, shows clearly the relationships between variables, between samples, and between variables and samples, on a single diagram. The technique has been applied successfully to distinguish geochemical anomalies from background, to recognize types of anomalies, to study the geochemical characteristics of deposits, etc. Several practical examples presented in this paper show that the method can be used to classify elements and samples into meaningful groups, and that the geochemical character of each sample group can be interpreted in terms of the corresponding element group. The results are consistent with the geological situations at hand, and are usually more elucidating than other classification methods.
[1]
C. Banfield.
Geological Factor Analysis (Methods in Geomathematics 1)
,
1977
.
[2]
Gilbert Saporta,et al.
L'analyse des données
,
1981
.
[3]
M. Hill.
Correspondence Analysis: A Neglected Multivariate Method
,
1974
.
[4]
F. Valenchon.
The use of correspondence analysis in geochemistry
,
1982
.
[5]
G. J. S. Govett,et al.
Handbook of exploration geochemistry
,
1981
.
[6]
Paul M. Mather.
Computational Methods of Multivariate Analysis in Physical Geography
,
1976
.
[7]
Richard Darling,et al.
Progresses in R- and Q-Mode Analysis: Correspondence Analysis and its Application to the Study of Geological Processes
,
1974
.