Prediction of mechanical properties of dental composite materials using machine learning algorithms

Abstract The durability of dental materials, when used in the mouth, is determined by their mechanical qualities. Composite resins are frequently used in dental restorations. Flexural tests and Vickers micro‐hardness tests on selected dental composite materials were performed in a universal testing machine (ASTM D790‐10 standard) and Vickers micro‐hardness tester (ASTM E384‐11e1standard). In this study, four different dental composite material samples are employed. The samples are dipped in a chewing tobacco solution for a few days before being removed and put through the tests. Also in this work, four different machine learning models were tested to see how well they could analyze the mechanical characteristics of dental composite materials when submerged in a chewing tobacco solution. For predicting the mechanical properties of dental composite specimens, four distinct machine‐learning models (extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), random forest, and k‐nearest neighbors (KNN) have been selected. AdaBoost machine learning model yields a coefficient of regression value of 0.9903 in predicting the flexural strength, whereas the XGBoost model gives a coefficient of regression value of 0.9890 in predicting the Vickers hardness distinctly better than the other models.