Composition Analysis and Identification of Ancient Glass Products Based on Hierarchical Clustering

Abstract

The ancient glass is very easy to be affected and weathered by the burial environment. During the weathering process, there will be a lot of exchange between internal elements and environmental elements, and the composition of glass will also change in proportion, which will affect the judgment of its category; This paper mainly deals with data based on hierarchical clustering analysis and the optimization model of hierarchical clustering analysis.First, the chi square test is used to analyze the data of the four variables. Based on whether the P significance value is less than 0.05, it is further judged that there is no significant difference between the surface weathering and the color and decoration, but only significant difference with the type of glass; Further, the corresponding analysis of glass type and whether there is weathering on the surface is made to draw a more specific conclusion; Count the difference of chemical composition addition of different types of glass with or without weathering, and then systematically describe and count the difference of values to obtain a histogram for observing the difference before and after the change; According to the data of weathered points and unweathered points, the data of unweathered points are extracted, and the average value of each chemical composition of weathered and unweathered points is calculated. The unweathered points do not show weathering of chemical composition on the time axis, so the average value is compared as the content of chemical composition before weathering.