Composition analysis and identification of ancient glass products based on correlation analysis

Aiming at the classification and weathering of ancient glass artifacts, this paper proposes a method based on multivariate statistical analysis and machine learning. First, the correlation between surface weathering and glass type, pattern and color was analyzed by Spearman correlation coefficient, and it was found that the correlation between glass type and surface weathering was the highest (correlation coefficient was -0.344). Secondly, a relationship model between chemical composition, glass type and surface weathering was established by multivariate linear regression, and decision tree and K-means clustering algorithms were used to classify and subclassify glass artifacts. The experimental results show that the proposed method performs well in classification accuracy and model stability. Finally, the principal components of chemical composition were extracted by principal component analysis, which provides a theoretical basis for the classification of high potassium glass and lead-barium glass. This study provides new ideas and methods for the protection and research of ancient glass artifacts, which has important academic value and application prospects.