Reducing the dimensionality of the set of predictors for predicting the solubility of impurities in copper-based solid solutions

Abstract

This study investigates the use of dimensionality reduction techniques to improve the accuracy of impurity solubility prediction in copper-based alloys. Principal Component Analysis (PCA) and correlation based distance metrics are combined to identify the most influential predictors and classify solubility behaviour. The proposed approach offers a robust foundation for the development of predictive models in the field of alloy design.