Welcome to Francis Academic Press

Academic Journal of Computing & Information Science, 2023, 6(7); doi: 10.25236/AJCIS.2023.060724.

Classification Model of Ancient Glass Artifacts Using Fisher Discriminant Analysis


Kaikai Kang, Weijie Wu, Xianchun He

Corresponding Author:
Weijie Wu

College of Computer and Data Science/College of Software, Fuzhou University, Fuzhou, 350108, China


Ancient glass artifacts are highly susceptible to weathering in burial environments, which can cause changes in their chemical composition. To address the problem of identifying and analysing the composition of ancient glass artifacts, this paper proposes a classification and identification model based on the Fisher discriminant method. Firstly, considering that the proportion of components in glass can change during the weathering process, which can affect the correct identification of its category, this paper predicts the data before weathering based on the normal distribution law and performs closed operations to ensure that the sum of the component data is 100%. Secondly, through variance analysis and kernel density estimation, the factors affecting the classification are selected, and a Fisher linear discriminant model is established. Finally, cross-validation is performed by applying Fisher linear discriminant to both the training and testing data, and the discriminant results of the two groups of data are found to be consistent with the actual situation, with a discrimination accuracy rate of 100%.


Identification of Ancient Glass Artifacts, Fisher Linear Discriminant, Kernel Density Estimation

Cite This Paper

Kaikai Kang, Weijie Wu, Xianchun He. Classification Model of Ancient Glass Artifacts Using Fisher Discriminant Analysis. Academic Journal of Computing & Information Science (2023), Vol. 6, Issue 7: 182-187. https://doi.org/10.25236/AJCIS.2023.060724.


[1] Yin Yulong. Analysis of ancient glass composition through correlation prediction [J]. Contemporary Chemical Research, 2023, (126, 1): 122-126.

[2] Jia Dawei, Wu Ziyan, He Xiang. Multivariate correlated kernel density estimation method for probabilistic seismic demand analysis under multidimensional performance limit states [J]. Journal of Vibration Engineering, 2022, 35(6): 1299-1310.

[3] Mao Shanshan, Ma Shengguo, Wei Benhai, et al. Delay estimation of power line communication network based on kernel density [J]. Electronic Design Engineering, 2023, 31(1): 133-137+142.

[4] Gui Lin, Yang Jianbo, Huang Yuanshuai. Value of logistic regression and Fisher linear discriminant analysis models in differential diagnosis of ovarian tumors [J]. Chongqing Medicine, 2018, 47(6): 800-802.

[5] Madaki U Y, Singh V V, Chiwa M D. Assessment of Diabetic Patients Among Adults in Maiduguri, Using Multivariate Discriminant Model[J]. Biostatistics and Biometrics Open Access Journal, 2019, 9.