Machine learning application in evaluation of graphite plates.

In the rapidly evolving field of emerging technologies, machine learning has the ability to advance the material science field significantly. Although there are many techniques to determine the material properties and behavior, we used machine learning as a key technique, which allows the machine to make data-driven decisions for carrying out certain tasks. There exist tremendous traditional methods and computational data modelling systems that consume a lot of time and require a lot of resources. Thus, it is mandatory to develop a mathematical modelling system that reduces time and works more efficiently than usual. Here we build upon the machine learning approaches to predict the nanoscopic changes, which include strain instabilities and variations in the structural changes of graphite. In this paper, we primarily focus on the material behaviour of graphite plates and use deep learning techniques to investigate the microstructural material properties and compare them to other techniques for predicting the full results. Here we have obtained nearly 8000 organized values from 29 different data sets. Raman spectroscopy was used to conduct experiments on graphite material for data sets. We used K-Nearest Neighbor (KNN) clustering algorithm to determine the changes in graphite material properties. [ABSTRACT FROM AUTHOR]