Xiaohan Deng1, Tangliang Wang2
1College of Science, Chongqing University of Technology, Chongqing, 400054, China
2College of Mechanical Engineering, Chongqing University of Technology, Chongqing, 400054, China
The surface weathering of ancient glass can be distinguished from the type of ancient glass by chemical composition as well as color and ornamentation. This paper constructs mathematical models to analyze the relationship between surface weathering and glass type, color and ornamentation based on different chemical compositions and characteristic data of color and ornamentation in ancient glass, explores the statistical law of chemical composition content with and without weathering, and constructs an effective classification model to realize glass type classification. This paper solves the relationship between glass type, decoration and color and their weathering degree; analyzes the statistical law of chemical composition content with and without weathering on the surface of two types of glass; derives the prediction formula and predicts the chemical composition content of two types of glass before weathering.
Random Forest Model; Multiple Regression Model; High Potassium Glass; Lead Barium Glass
Xiaohan Deng, Tangliang Wang. Multiple regression-based glass composition prediction and statistics. Academic Journal of Mathematical Sciences (2022) Vol. 3, Issue 2: 57-62. https://doi.org/10.25236/AJMS.2022.030209.
 Wang Chengxian, Tao Ying. Weathering of silicate glasses [J]. Journal of Silicates, 2003, 31((1): 78-85.
 Lu Yahui. Bayesian classification model based on random forest feature selection and application [C]. North China University of Water Resources and Hydropower, 2017.
 Chen, Neng-Mei; Liu, Xiao-Jing. Research on the classification model of channel scene based on random forest algorithm [J]. Journal of Chongqing University of Technology (Natural Sciences), 2017, 31(04):134-140.
 Zhu Yixingbo. Research and application of automotive user portrait based on improved K-mean clustering algorithm[C]. Jilin University, 2022.
 Gao A-F. Research on tillage depth prediction model based on improved random forest [C]. Changchun University of Technology, 2022.