Abstract
In recent years, the Internet has developed rapidly, and in the face of thousands of data and information, it has become very critical for users to find the information that is of high value to them in the mass of information, and the recommendation system is one of the most effective ways to solve this information overload phenomenon. In this paper, the current movie recommendation algorithm is improved by using an item-based collaborative filtering algorithm for the similarity measure of items in the item-based recommendation process; In the recommendation process, two more applicable recommendation methods are considered: collaborative filtering content-based recommendation and matrix decomposition-based recommendation. It saves users time in searching, viewing and filtering, while discovering information about their potential movie preferences.
Publisher
Darcy & Roy Press Co. Ltd.
Reference8 articles.
1. Hua-fu Duan. Application of matrix decomposition techniques considering time effects in recommender systems[J]. Microcomputer applications, 2013, 29(03): 53-55+64.
2. Liang-jun Liu, Liu Yang. Design and implementation of movie recommendation system[J]. Internet of things technology, 2021, 11(03): 86-88+92.
3. Zong-yan Chen. Collaborative filtering recommendation algorithm for sparse data optimization[D]. Nanjing University of Posts and Telecommunications, 2017.
4. Deuk Hee Park, Hyea Kyeong Kim, Il Young Choi, Jae Kyeong Kim. A literature review and classification of recommender systems research[J]. Expert Systems with Applications, 2012, 39(11).
5. Yao-ning Fang, Yun-fei Guo, Xue-tao Ding, Ju-long Lan. An improved recommendation algorithm for singular value decomposition based on local structure[J]. Journal of Electronics and Information, 2013, 35(06): 1284-1289.