Prediction of Hardness and Fracture Toughness in Liquid-Phase-Sintered Alumina System Using Gaussian Process Regression and Minimax Probability Machine Regression. (2018)

• Record Type: Journal Article
• Authors: Gopinath, K.G.S.; Pal, Soumen; Tambe, Pankaj
• Is Part Of: Materials today. Volume 5:Number 5(2018)Part 2
• Page Start: 12223
• Page End: 12232
• Publication Date: 2018
• Subjects: Hardness -- Fracture Toughness -- Gaussian Process Regression -- Minimax Probability Machine Regression -- LPS alumina; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2018.02.199
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: For Liquid Phase Sintered alumina, Hardness and Fracture Toughness are very important parameters for measure of wear and tear strength in ceramics. In this work, we apply Gaussian Process Regression and Minimax Probability Machine Regression for predicting hardness and fracture toughness. The performance of Gaussian Process Regression and Minimax Probability Machine Regression for predicting the hardness and fracture toughness is measured by using the correlation of coefficient or robustness (R). The Gaussian Process Regression outperforms over Minimax Probability Machine Regression and the Gaussian Process Regression is much more suitable for predicting the parameters of Liquid Phase Sintered alumina.