Component Analysis and Identification of Ancient Glass Products Based on Correlation Analysis

Abstract

The Silk Road was an important channel for cultural exchanges between ancient China and the West, and glass was a precious material evidence of early trade exchanges. In order to explore the changes of relevant properties and chemical compositions of these glass relics before and after weathering, it is necessary to calculate their correlation degree and predict the changes of compositions. In this paper, the correlation and correlation between the components are calculated by establishing the grey correlation analysis method, independent sample T test method, random forest algorithm, K-means algorithm and other models, so as to analyze and predict the correlation properties and chemical components of glass relics before and after weathering.first, preprocess the data, remove the invalid data, and take the glass type, decoration and color as the comparison sequence. Whether the cultural relic surface is differentiated into a reference sequence, use the gray correlation analysis method to calculate that the gray correlation between the glass type and whether it is differentiated is the largest, which is 0.7939, indicating that the glass type is most closely related to whether it is weathered. According to the type of glass, the average, median and standard deviation are calculated, and the chemical composition with strong correlation with weathering is determined by independent sample t-test. The prediction results are obtained by predicting the principal components of weathering points.