| The rapid development of computer network has greatly promoted the process of educational informatization,and accelerated the growth of educational information data on the Internet.It's a tremendous challenge to the relevant personnel of the education management department to classify,manage and evaluate these resources of education news and detection of sensitive information timely.Aiming at the problem of emotional evaluation of educational news,in this paper,education news is classified into positive news and negative news,and a Naive Bayes education news classification method based on emotion dictionary is proposed.The main research work of this paper is as follows:(1)Education's emotional dictionary is built and expanded.First of all,the basic emotional dictionary is organized by the Chinese emotional polarity dictionary of National Taiwan University and the Chinese emotional vocabulary ontology library of Dalian University of Technology.Then,with education news as the application background,the emotional vocabulary related to education is summarized and sorted on the basis of the basic emotion dictionary,and the education news emotion dictionary is constructed.Finally,the smoothing SO-PMI algorithm is used to expand the emotional dictionary of educational news.A highly effective emotional dictionary for education news text has been formed,and this emotional dictionary is used for education news emotion classification.(2)A Naive Bayes education classification method based on emotion dictionary is proposed.Firstly,data preprocessing of education news data source is conducted.Secondly,the Chinese text representation of educational news through the vector space model,the emotional dictionary of education news is used as the basis for feature selection,and the TF-IDF is used to perform feature weight calculations.Finally,the education news classifier constructed by the Naive Bayesian classification algorithm divide the education news into two categories:positive news and negative news.The experiments show that the Naive Bayes education news classification method based on emotional dictionary has higher accuracy and recall rate.(3)An education news emotional evaluation classification system was designed and implemented.Aiming at the needs of the education management department of a city,this paper uses the Naive Bayesian classification method based on emotional dictionaries to design and implement an education news emotional evaluation classification system.This system is practical and friendly,with the function of the education news text classification,statistical analysis,hot word cloud,keyword search and so on.At present,the emotional evaluation classification system of educational news has been put into use,and its performance is stable and user experience is good. |
PDF Full Text Request