Research On Text Classification Based On Improved Convolutional Neural Network

With the rapid development of Internet technology,the rapid growth of text data has brought about.Under this growth trend,the use of deep learning models to classify texts has attracted the attention of researchers.In recent years,Convolutional Neural Networks(Convolution Neural Network,CNN)as an important network model in deep learning can use its own advantages to classify large-scale text data,indicating that CNN has great application value in the field of text classification.This paper mainly studies the CNN-based text classification problem.First,a CNN-based text classification model is constructed.After comparing the experimental results,the CNN-based text classification model uses the ReLU activation function and the maximum pooling method.The problem of over-fitting and time-consuming training of the text classification model of the text,built a text classification model(Sparse dropout constrained convolutional neural network,SDCNN)based on sparse Dropout convolutional neural network.In order to verify the classification performance of the CNN model and SDCNN model on text data,this article uses the English Reuters-21578 data set for experiments.The experimental results show that the classification accuracy and recall rate of the CNN model constructed in this paper are better than the KNN model and BP model,SVM model and AE model.Compared to the CNN model,the SDCNN model not only solves the problems of overfitting and training time-consuming,but also improves the accuracy of text classification.