Component Analysis and Identification of Glass Products Based on Support Vector Machine Comprehensive Classification Model

Abstract

It is of great significance to study the ancient Silk Road to analyze and identify the types and components of glass products. In this paper, support vector machine (SVM) method is used to study the statistical rule and internal relation between different chemical components of ancient glass products and how to classify ancient glass products. Firstly, the high-potassium glass and lead-barium glass were further divided into subclasses, and then the K-means cluster analysis was carried out. After many fitting repeats, the optimal number of classes was obtained as 5, and the chemical composition characteristics of the classified types were quantitatively analyzed and their ranges were divided. Finally, the five glass types were named as PbO- BaO- SiO2, PbO (~)- BaO- SiO2, K2O- SiO2- Al2O3, PbO- BaO (~)- SiO2- CuO and K2O- SiO2 (~).Then the classification criteria of each category are given on the basis of K-means clustering, and then combined with the support vector machine model, the unknown categories of glass artifacts are classified by machine learning, and finally the final results are obtained by fitting the five ancient glass artifact subcategories based on multiple chemical compositions. And the result probability is predicted from the 5 subclasses, and the maximum probability is taken as the final predicted category.