Analysis and identification of glass products based on fisher linear discriminant analysis and logistic regression for dichotomous classification

Abstract

This paper focuses on establishing multiple linear regression models, correlation analysis, fisher linear discriminant analysis, logistic regression to deal with dichotomous classification, sensitivity analysis, systematic clustering and gray correlation analysis to solve the problem of compositional analysis and identification of ancient glass products. The subclasses of each category of glass are divided, and the division results are derived, so that the rationality and sensitivity of the classification results can be analyzed, and finally the classification laws of high potassium glass and lead-barium glass are derived. Analysis The chemical composition of unknown glass artifacts, predict their glass types, and perform sensitivity analysis on the predicted results.