Oral English Course Online Recommendation System Based on Support Vector Machine

Abstract

Aiming at the problems of existing oral English online teaching system that the course recommendation model is not accurate enough, which leads to the low accuracy of system recommendation. This article improves the shortcomings of collaborative filtering algorithms and designs an online course recommendation system based on KNN-SVM. The course recommendation model is based on the similarity between the curriculum and student behavior recommended lesson list which is generated from the student, the history of collecting data, and the interaction of the user. Then, KNN is used to perform secondary clustering on user and topic matrices, and support vector machine cross validation is used for final classificationThe average accuracy of the oral English online education course recommendation system in this paper is 57.926%, 45.391% and 44.399%, respectively, compared with the other two kinds of course recommendation systems, indicating that the designed oral English online education course recommendation system has better effect after combining artificial intelligence technology.