Artificial intelligence technology is used in various fields as the fourth industrial revolution begins. In this study, the surface cracks of the continuous casting process are predicted by binary classification using neural network algorithm which is one of the lesson learning of machine learning. For this purpose, binary classification was performed based on the data extracted from the field. Inputs were MCI (mold crack index), SCI (Strand Crack Index) and GI (Gap Index). In addition, the sigmoid function was used as an activation function, and the softmax function was applied to the output layer to improve the prediction layer. The number of hidden nodes is increased by 1 to 20 for each layer and the number of hidden nodes with low cross entropy error (CEE) and low prediction error is used. Based on this, we predicted surface cracking in a continuous casting process.