Recommendation of Online Educational Resources Based on Neural Network

摘要

In recent years, online education is developing rapidly. However, it is difficult for students to obtain suitable content from massive online educational resources. Therefore, it is of great significance to recommend personalized educational resources to students. How to effectively model the characteristics of students and educational resources, and to explore the deep relationship between them is one of the core challenges of the recommendation model. In this paper, we propose an educational resources recommendation model based on neural network. Based on the data of Spark Space, the online learning platform of Beijing University of Posts and Telecommunications, we first construct a knowledge concept maps according to the knowledge structure of the course. And then we combine the learning style scale and Bloom’s taxonomy theory to construct the characteristic attributes of students and education resources. Finally, we use a multilayer perceptron network to make personalized recommendations to students. Experiments have proved that the recommendation model proposed achieves a great recommendation result.