Research On Text Classification Method Based On Deep Learning

In recent years,with the rapid development of network technology,the number of network users has shown explosive growth.Every day,there are tens of thousands of short texts coming into being.These short texts include weather,politics,economics,culture,reviews of movies,and so on.How to classify these short texts effectively so as to better serve the network users has become the key.And the technology of text classification can provide an effective way to solve the problem.This paper studies and designs a text classification method based on ensemble learning.The main work is as follows:Firstly,this paper introduces the traditional machine learning text classification methods and deep learning text classification methods.Among them,machine learning methods include: naive Bayes?K-nearest Neighbor(KNN),Decision Tree.Deep learning text classification methods include: Convolutional Neural Networks(CNN),Bidirectional Long Short Term Memory(Bi-LSTM),Convolutional-Long Short Term Memory(C-LSTM),Recurrent Convolutional Neural Networks(RCNN),Hierarchical Attention Networks(HAN).The advantages and disadvantages of these classification algorithms are analyzed.Secondly,this paper elaborates on the relevant theoretical knowledge of ensemble learning.On this basis,the deep learning algorithm with the above five classification effects is designed and implemented as the base classifier.The ensemble method of Bagging and Stacking is used to carry out experiments in turn,which improves the classification accuracy of the base classifier.Then the paper uses different combination strategies to test.The experimental results show that the classification accuracy of the ensemble learning method is higher than that of the base classifier model.Pairwise model combination experiments are carried out to verifies the contribution of a single model to short text classification.Finally,on the basis of summarizing the whole text and summarizing the experimental results,the prospect of the next step is put forward.