Research on the Composition Analysis and Identification of Ancient Glass Products Based on Decision Tree and K-means Clustering Algorithm

Abstract

Ancient glass is susceptible to weathering by environmental influences, resulting in changes in its chemical composition and proportions, which affects the correct judgment of its category by archaeologists. In order to correctly analyze and identify the composition of ancient glass artifacts, firstly, this paper establishes a decision tree model to derive the basis for determining the type of glass artifacts. Secondly, the aggregation coefficients of high potassium glass and lead-barium glass were plotted separately in this paper, and the K-value of clustering was determined to be 2 according to the elbow rule. After using principal component analysis to reduce the dimensionality of the chemical composition content data, the K-means clustering algorithm was used to classify the subclasses of high potassium glass and lead-barium glass respectively. Finally, the established model was used to identify the chemical composition and its type of unknown class of glass artifacts.