SELECTION OF MACHINE LEARNING ALGORITHMS FOR NANOCOMPOSITE ZA-27 MATERIAL TRANSFER PREDICTION

Abstract

This study explores the use of machine learning algorithms in predicting material transfer in tribological contacts. The results of the analysis indicate that the machine learning models can accurately predict the occurrence of material transfer with a high degree of accuracy. The Gradient Boosting Classifier algorithm was found to outperform other algorithms in terms of predictive accuracy. The study's practical implications suggest that machine learning can be an effective tool for predicting and preventing material transfer, leading to increased system reliability and durability. The findings highlight the importance of domain-specific expertise in selecting appropriate algorithms and input features. One limitation of the study is that it focused only on material transfer and did not consider other important factors such as wear and friction. Future research could investigate the use of machine learning algorithms in predicting wear and friction in tribological systems.