摘要
古代玻璃制品极易受埋藏环境的影响而风化,导致玻璃文物成分比例发生变化,从而影响对其类别的正确判断。文章针对古代玻璃制品的成分分析与鉴别问题,使用卡方检验、K-Means均值聚类、分层聚类、灰色关联等方法,构建卡方模型、BP神经网络模型、K-Means均值聚类模型、灰色关联分析等,运用Matlab、SPSS、Python等软件编程,研究了玻璃类型高钾与铅钡的分类依据,并对不同类别玻璃按照化学成分进行亚分类,分析了其化学成分的统计规律与关联关系。最后利用单因素方差分析法,检验了模型的合理性。
Ancient glass products are easily weathered under the influence of burial environment,which leads to the change of the composition proportion of glass cultural relics,thus affecting the correct judgment of their categories.Aiming at the problem of composition analysis and identification of ancient glass products,this paper uses Chi-square test,K-Means clustering,hierarchical clustering,grey correlation and other methods to build Chi-square model,BP neural network model,K-means clustering model,grey correlation analysis,and uses Matlab,SPSS,Python and other software programming.The classification basis of high potassium and lead barium glass types was studied,and different types of glass were subclassified according to their chemical compositions,and the statistical rule and correlation of their chemical compositions were analyzed.Finally,the rationality of the model is tested by single factor analysis of variance.
作者
徐惠
胡珊
姚旭敏
储昭顺
XU Hui;HU Shan;YAO Xu-min;CHU Zhao-shun(School of Statistics and Applied Mathematics,Anhui University of Finance and Economics,Bengbu,Anhui,233030,China)
出处
《新疆师范大学学报（自然科学版）》
2023年第3期66-73,96,共9页
Journal of Xinjiang Normal University(Natural Sciences Edition)
基金
国家社会科学基金年度项目(22BTJ048)。
