Research And Implementation Of Chinese Text Classification Algorithm Based On Machine Learning

With the rapid development of big data era,the amount of text information data increases sharply.In order to obtain valuable information and improve the efficiency of information acquisition,it is necessary to classify these text information.Therefore,the research and implementation of text classification system is of great significance.News text is an important part of text information and an important way for people to get information.Based on the classification of news texts,this paper improves the current text classification algorithm,explains the design and implementation of the text classification system by taking the news text classification system as an example,and completes the research on the text classification algorithm.Based on the background of big data era and the development of classification technology,the main work includes the following aspects:First,this paper conducts a survey on different text classification methods.This paper understands the history through reading literature materials and simple experimental exploration and analyzes the advantages and disadvantages of Bayes,KNN,SVM,decision tree,Fasttext,CNN and other classification algorithms.Secondly,a k-Bayes classification algorithm combining the distribution of feature words was proposed,which increased the weight of feature words with prominent significance and improved the efficiency of text classification.Thirdly,feature extraction operation of attention layer is added to further extract characteristic words in the convolutional neural network,which conforms to people's understanding of classification,which has a good effect on improving the accuracy of text classification.Fourthly,this paper designs and implements a news text classification system,which can display intuitive classification results for users.Through experimental verification,the classification algorithm in this paper has indeed improved the classification accuracy.