Research on Composition Analysis and Identification of Ancient Glass Products

Abstract

Glassware was the witness of the ancient Silk Road trade. The paper focuses on the component analysis and identification methods of ancient glass products. The decision tree model was established and analyzed to obtain the classification rules of high potassium glass and lead barium glass. Based on the cluster analysis method, appropriate chemical elements were selected for each type to conduct cluster analysis. The result showed that high potassium glass and unweathered lead barium glass can be classified by their content of silicon dioxide. However, weathered lead barium glass needed to be classified by their content of lead oxide. The random forest regression model was used to classify glass cultural relics by their chemical elements. The chemical element data of glass cultural relics with known types were used as samples, and the chemical element data of glass cultural relics with unknown types were inputted into the trained model. Finally, the types of glass cultural relics were predicted and sensitivity analysis was conducted.