Research on the Composition of Ancient Glass Products Based on Statistical Analysis Method
Spearman correlation coefficient and chi-square test were used to analyze the composition of ancient glass,and the correlation and difference of glass type,decoration,color and surface weathering were obtained.The median,mean,standard deviation,variance,kurtosis and skewness were compared and analyzed for the data of weaved glass products,and the weighted average prediction model was used to obtain the prediction before weathering,and the identification method of glass type was given.Using the formula of sensitivity coefficient,the ratio of silica content plus or minus 10%and lead oxide content is obtained,which indicates that the slight change of lead oxide and silica content will not affect the category.With the help of correlation analysis and pivot table,it is concluded that high potassium glass and lead-barium glass have high content of silica,alumina,calcium oxide and copper oxide before weathering,and the chemical composition of potassium glass is weak after weathering,and the chemical composition of lead-barium glass is strong after weathering.
Spearman correlation coefficientChi-square testweighted average prediction modelglass type
NETL
NSTL
万方数据
