Semantic Analysis of Public Health Medical Issues Based on Convolution Neural Networks
Algorithm 2
The convolutional neural network analysis algorithm.
Input: Training dataset X and test dataset
Output: The results are predicted based on the trained model
Begin
For i = 1,…,N do
Step1. Sequential 1(conv1):
(1):
Input the data X into the convolution layer (the number of input channels is 1, the number of output channels is 10, convolution core size is 3, and convolution step is 1; after the data are filled, the convolutional kernel scans the data and extracts depth characteristics)
(2):
The function Relu() is used for the nonlinear processing of the data
(3):
Features of the data are extracted using maximum pooling, where the kernel_size is 2 and the step size is 2
Step2. Sequential 2(conv2):
(1):
The data obtained by Sequential1 is input (the input channel number is 10, the output channel number is 20, the convolutional kernel size is 3, and the convolutional step is 1, and the data’s feature is extracted again);
(2):
The function Relu() is used for the nonlinear processing of the data
(3):
Max Pooling is used to extract the features of the data, where the kernel_size is 2 and the step size is 2
(4):
The data obtained from Sequential2 is expanded and fed into the fully connected layer
end for
Step3. According to the above results, the loss function BCEWithLogitsLoss is used for multilabel classification of test data to obtain the classification results.
