Article Details

Abstract

Aiming at the error problem of similarity calculation in Collaborative Filtering recommendation algorithm, this paper proposes a similarity calculation method with correction factor and applies it to movie recommendation algorithm. Firstly, the movie information data is reduced in dimension and clustered by K-means algorithm to improve the efficiency of the algorithm. Secondly, the correction factor is introduced, and the movie similarity matrix is calculated by combining the Pearson correlation coefficient. Finally, according to the similarity matrix, the user's prediction score for the unrated movie is calculated to complete the recommendation. The experimental results show that compared with the traditional Collaborative Filtering recommendation algorithm, the efficiency of this algorithm is improved by 62.4%. The Mean Absolute Error of the optimized similarity algorithm is 9.67% lower than that of the traditional similarity algorithm.

Keywords

similarity matrix; correction factor; Principal Component Analysis; clustering; Collaborative Filtering

References

1. [1] Yu Meng, He Wentao, Zhou Xuchuan, etc Review of Recommendation Systems [J]. Computer Applications, 2022, 42 (6): 1898-1913
2. [2] Yu Saisai Research on prediction and recommendation of personalized movie behavior based on Internet users [D]. Linyi: Linyi University, 2024
3. [3] Zhao Junyi, Zhuang Fuzhen, Ao Xiang, etc Overview of Collaborative Filtering Recommendation Systems [J]. Journal of Information Security, 2021, 6 (5): 17-34
4. [4] Yuan Quan, Cheng Zhenhua, Jiang Yang Research on Movie Recommendation Algorithm Based on Knowledge Graph and Collaborative Filtering [J]. Computer Engineering and Science, 2020, 42 (4): 714-721
5. [5] Fu Fang Application of Fusion Clustering and Improved Similarity Collaborative Filtering in Movie Recommendation System [D]. Chongqing: Chongqing University of Technology, 2024
6. [6] WANG C，BLEI D M. Collaborative Topic Modeling for Recommending Scientific Articles [C]//Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining.New York：ACM，2011：448-456.
7. [7] HERLOCKER J L，KONSTAN J A，TERVEEN L G，et al. Evaluating Collaborative Filtering Recommender Systems [J].ACM Transactions on Information Systems (TOIS)，2004，22（1）：5-53.
8. [8] Wang Cheng Collaborative recommendation algorithm based on deep feature fusion [J]. Journal of Nanjing University of Science and Technology, 2024, 48 (4): 460-468
9. [9] Gu Yiran, Zhang Yuanzhi, Yang Haigen Movie recommendation algorithm based on movie attributes and interactive information [J]. Journal of Nanjing University of Science and Technology, 2022, 46 (2): 177-184
10. [10] Zeng Zhaowei, Cao Jian Recommendation of Data Analysis Service Process Model [J]. Small Microcomputer Systems, 2019, 40 (7): 1374-1379

Funding

Guangdong Province Higher Vocational Education Teaching Quality and Teaching Reform Project (2023JG048); Guangdong Provincial Education Science Planning Project (2022GXJK552); Guangdong Province 2022 Higher Education Research Platform and Key Project Areas Special Project (2022ZDZX1067)

About the authors

Chen Ying (1979-), female, Han ethnicity, from Jieyang, Guangdong, associate professor, undergraduate, research direction: new data management, information technology; Shenyang (1978-), male, Han ethnicity, from Anqing, Anhui Province, lecturer, doctoral student, research direction: network technology, big data integration management.