Higher Education Online Courses Personalized Recommendation Algorithm Based on Score and Attributes

Wenbin Wu,Bixi Wang,Wenfeng Zheng,Yan Liu,Lirong Yin
DOI: https://doi.org/10.1088/1742-6596/1673/1/012025
2020-11-01
Journal of Physics: Conference Series
Abstract:Abstract Education through the online platform has gained popularity. While the online course keeps increasing, its benefit does not increase in a proportional way. Personalized recommendation systems can well solve this problem by mining users’ interests and preferences. Aiming at the cold start problem, this paper improves the collaborative filtering algorithm by combining the user score and project attribute characteristics of the online course platform. The network communication technology is used to obtain the user ratings and project attribute data to verify the feasibility of the recommendation algorithm, adjust the parameters in the model, and compare the accuracy of the algorithm. The designed algorithm can provide accurate and rapid personalized recommendation services, which is convenient for users and conducive to the development of the platform. The improved algorithm solves the cold start problem compared with the traditional algorithm with a significantly improved prediction accuracy. The scheme can also be modified according to the changes in user preferences and can achieve good real-time recommendation effect.
What problem does this paper attempt to address?