A method of educational news classification based on emotional dictionary

Abstract

Aiming at the problem of educational news classification, this paper proposes a method of educational news classification based on emotional dictionary. This paper takes educational news as the application background. First, this paper summarizes the common sentiment dictionary and the education-related sentiment words of manually arranged. Then, this paper uses the Laplace smoothing SO-PMI algorithm to expand the emotional dictionary of educational news. Finally, the whole news is divided into several clauses, from which emotional key words in education news is required by combining emotional dictionary, and calculate the weights of each clause according to corresponding rules and get the weights sum of each clause, and we get the emotional tendencies of the whole education news in order to identify the positive and negative classification of educational news. Experimental results show that the method of educational news classification based on emotional news has higher precision and recall rate.