Prediction of quality material using machine learning algorithm performance

This study aims to develop a machine learning approach based on a material quality evaluation system (diagnosis). The proposed method is a custom implementation of the Naïve Bayes algorithm. The specimens consisted of four mortar variants with categories M, S, N, and O, tested at different ages of 3, 7, 14, 21, and 28 days, respectively. The pulse velocity results were analyzed and validated with a machine learning approach. The introduction of machine learning into this system facilitates the detection of experimental test results through ultrasonic wave propagation speed (UPV). Applying code to models proves that machine learning can test, evaluate, and interpret results.