PRINCIPAL COMPONENT ANALYSIS OF SPATIAL DATA (PREPRINT IBRAE-2001-18)

Preprint IBRAE-2001-18

Parkin R., Kanevski M., Maignan M., Raspa G., Hakamata T., Savelieva E. Kalantarov E.

This work presents review of various Principal Component Analysis (PCA) methods and application PCA for the spatial prediction of concentration of metals using software “Multigeo”. Principal component analysis is one of the most popular techniques for processing, compressing and visualizing data, although its effectiveness is limited by its global linearity. The technique is illustrated using the real data on Geneva Lake sediments contamination and Japanese soil contamination by heavy metals.   

Bibliographical reference

Parkin R., Kanevski M., Maignan M., Raspa G., Hakamata T., Savelieva E. Kalantarov E. PRINCIPAL COMPONENT ANALYSIS OF SPATIAL DATA. Preprint IBRAE-2001-18. Moscow: Nuclear Safety Institute RAS, 2001. 21 p. — Refs.: 41 items.