Application of Text Classification Method Based on Depth Learning in Email Handwriting Analysis

Natural Language Processing is an important direction in the fields of computer science and artificial intelligence. It has a wide range of applications, such as text classification, machine translation, emotional analysis, etc. where text classification is typically characterized of capturing the language features of e-mails to identify the authors. In this paper, mathematic models are set up and compared based on the traditional text classification method and the Deep Learning text classification method respectively. Finally, the following results are obtained: The accuracy of the model using the improved network architecture based on TextCNN reaches 87% on the training set and 88.97% on the test set. The accuracy of using the improved network architecture based on CLSTM reaches 90% on the training set and 88.34% on the test set. The accuracy of the new network architecture based on recurrent neural network (RNN) and convolutional neural network (CNN) has reached 92.32% on the training set and 92.80% on the test set. Using integrated learning to integrate the model, the accuracy rate is 94% on the training set and 92.36% on the test. The model based on SVM are tested with the test set to get the accuracy rate 60.44%.