Prediction of Weight Percentage Alumina and Pore Volume Fraction in Bio-Ceramics Using Gaussian Process Regression and Minimax Probability Machine Regression

Abstract

In Bio-ceramics, the alumina weight percentage and pore volume fraction play a vital role for its biocompatibility in human body. There are many experimental methods which are employed for achieving the required quality in it. In this work, for preparation of Al2O3/SiC ceramic cake, the amount of Silicon Carbide (SiC) is taken as input parameter. The weight percentage Alumina and pore volume fraction are taken as output parameters. Two machine learning models such as Gaussian Process Regression (GPR) and Minimax Probability Machine Regression (MPMR) are applied for predicting the above two output parameters. The performance of the above two models are compared. The Gaussian Process Regression outperforms the Minimax Probability Machine Regression marginally and the result of the Gaussian is encouraging for predicting the above two outputs.