“…The automatic encoder is an unsupervised selflearning mode with three fully connected neural network structures, including the input layer, the hidden layer, and the output layer. The encoder can map the input data from the high-dimensional space into the code in the low-dimensional space, and the decoder reconstructs the output data in the feature layer into the original input data, and obtains the effective features of the input data from the hidden layer through the encoding process and decoding process [16]. In this research, the constructed automatic encoder structure is shown in figure 2.…”