WU Jun, YIN Li, ZHANG Maolin, WU Junming, LI Qijiang

(Jingdezhen Ceramic Institute, Jingdezhen 333001, Jiangxi, China)

Abstract: The chemical composition test results of the body of Jingdezhen imitated Longquan and Longquan celadon were studied by multivariate statistical discriminant analysis and artificial neural networks for their respective provenance and chronology. The differences and applicability of the two methods were discussed. Results show that as the data of ancient ceramic element composition couldn’t fully meet its requirements, the accuracy of the multivariate statistical discriminant analysis is lower than that of the artificial neural networks, which means the artificial neural networks is more suitable for ancient ceramic provenance and chronology determination.

Key words: ancient ceramics; provenance and chronology; BP artificial neural network; multivariate statistical analysis; discriminant analysis