Principal component analysis and artificial neural networks applied to the classification of Chinese pottery of neolithic age

Volumetric analysis, as a simple, rapid, accurate and economic method, has been used in studying the chemical composition of Chinese neolithic age pottery. The major component analysis, principal component analysis (PCA) and artificial neural networks (ANNs) have been used to classify these potteries; the results show that they belong to three categories, the Yellow River Valley (YR) region, the Yangtse River Valley (YV) region and other region (OR). This work reveals that the ANN seems to be more suitable than PCA in classifying such archaeological samples. (C) 2000 Elsevier Science B.V. All rights reserved.