Classification and study of glass based on cluster analysis

Abstract

As one of the earliest human materials invented by mankind, glass products play a great role in the study of history. Because the ancient glass is easily affected by the buried environment, resulting in the weathering of the surface of the cultural relics, this paper classifies the cultural relics in order to better protect and study the ancient glass. In this paper, some data of weathered cultural relics are collected to establish a model for subclassification of cultural relics.This paper first conducted principal component analysis of two different types of glass and screened out different principal components of two types of glass. Then, according to the number of cluster categories (K value) obtained by the contour algorithm, combined with the obtained principal components, the cluster analysis is used to divide the subclasses, and the specific subclass division method and division results are obtained. The subclass characteristics are obtained by analyzing for each subclass data. For the established subclass classification, multiple cluster analysis was performed using changed K values to perform the sensitivity and rationality analysis of the model. The results show that the model established in this paper finds a good classification rule for some chemical substances of weathered cultural relics, gives a logical division method, and explores and compares the similarities and differences between different chemical components.