Welcome to Francis Academic Press

Academic Journal of Materials & Chemistry, 2023, 4(6); doi: 10.25236/AJMC.2023.040604.

Identification of Ancient Glass Based on Multiple Rinear Regression


Cui Wang

Corresponding Author:
Cui Wang

School of Management, Shandong University of Technology, Zibo, Shandong, 255000, China


The weathering of ancient glass under the influence of burial environment will result in the change of the composition ratio of glass. In order to identify the types of ancient glass products, this paper first establishes four multiple regression equations through a batch of existing weathered glass sample detection data, and obtains the chemical composition content rule of ancient glass surface weathering. Then, the difference between the absolute value of the predicted value and the actual value is the minimum as the basis for the identification of the unknown category of glass relics. Finally, the perceptron model optimized by particle swarm optimization is used as the auxiliary evidence to compare the results of the two, analyze the different categories, and finally get a more accurate answer.


Multiple linear regression, Perceptron model, Classification of ancient glass

Cite This Paper

Cui Wang. Identification of Ancient Glass Based on Multiple Rinear Regression. Academic Journal of Materials & Chemistry (2023) Vol. 4, Issue 6: 19-24. https://doi.org/10.25236/AJMC.2023.040604.


[1] Liu Song, Li Qinghui, Gan Fuxi. Influence of Surface Factors on Portable X-ray Fluorescence Analysis of ancient glass samples [J]. Spectroscopy and spectral analysis, 2011, 31(07): 1954-1959.

[2] Bethany Matthews, Bruce Arey, Carolyn Pearce, Albert Kruger. Characterization of Glass Alterations in Ancient Glass from Various Environments from Broborg, a Vitrified Swedish Hillfort [J]. Microscopy and Microanalysis, 2020, 26(S2).

[3] Feng Bailing. The Construction and application of the database of Ancient Glass Beads unearthed in China [D]. Northwest University, 2021. DOI: 10.27405/d.cnki.gxbdu. 2021.001676.

[4] Yin Yulong. Composition Analysis of ancient glass Products by Association Prediction [J]. Contemporary Chemical Industry Research, 2023, No. 126(01): 122-126.

[5] Lu Jiajia. Classification model of ancient glass based on integrated feature selection and random forest [J]. Journal of Ceramics, 2023, 51(04): 1060-1065. DOI: 10.14062/j.issn. 0454-5648.20220790.