Author(s)

Li Linze

Corresponding Author

Li Linze

ABSTRACT

Improving the recommendation performance of the recommendation system has been a very big challenge in the past, because both the accuracy of the recommendation results and the calculation time for calculating the recommendation results must be taken into account when making recommendations. Based on the above problems, in this paper we propose a recommendation algorithm based on deep learning, which uses deep learning methods to mine features of users and movies and train models to improve the accuracy of recommendation algorithms. At the same time, the features of users and movies are extracted through neural networks, rather than based on the user's rating matrix for movies, which solves the sparsity problem and cold start problem in the recommendation system. Finally, experiments are conducted on real data sets to verify the accuracy of the recommended algorithm.

KEYWORDS

Deep learning; movies; recommendation algorithm; cold start