Computational of Concrete Slump Model Based on H2O Deep Learning framework and Bagging to reduce Effects of Noise and Overfitting

Concrete mixture design for concrete slump test has many characteristics and mostly noisy. Such data will affect prediction of machine learning. This study aims to experiment on H2O Deep Learning framework Bagging noisy overfitting avoidance create the Slump Model. The consists cement, blast furnace slag, fly ash, water, superplasticizer, coarse aggregate, fine age, slump, compressive strength. A primary mixed using aggregate material from Merapi Volcano, hills Muntilan, Kalioro. was obtained Pamotan, Jepara, Semarang, Ungaran, Mojosongo Boyolali Central Java. cement Gresik Holcim product water Tembalang, Semarang. model with one input layer 7 neurons, hidden 20 output 1 neuron activation function TanH, parameter L1=1.0E-5, L2=0.0, max weight=10.0, epsilon=1.0E-8, rho=0.99, epoch=800 is able achieve RMSE 2.272. result shows that after introducing Bagging, error can be reduced up 2.5 approximately (50% lower) compared without Bagging. manually tested used evaluation. 0.568. Following this study, further research such as creating practicum equipment/ application software.