Open AccessCollaborative Filtering Based on a New Matrix Factorization Algorithm
Author(s) -
Yongjie Yan,
Hui Xie
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1650/2/022032
Subject(s) - collaborative filtering , computer science , cluster analysis , recommender system , matrix decomposition , spectral clustering , data mining , algorithm , process (computing) , quality (philosophy) , information retrieval , machine learning , philosophy , eigenvalues and eigenvectors , physics , epistemology , quantum mechanics , operating system
Online platforms such as Facebook, Netflix, and Amazon, provide users with information on products or services that matches their interests or backgrounds. As one of the most popular services over online platforms, recommendation system has attracted increasing research efforts recently. Among all the recommendation tasks, an important one is item recommendation over high spectral data. Existing recommendation techniques are not effective for handling users with diverse interests. In this paper, the sparsity problem of traditional recommendation algorithms, a new recommendation algorithm based on spectral clustering is proposed. The spectral clustering process can improve the efficiency of spectral clustering algorithm. Spectral clustering can be performed off-line, which will accelerate the speed of online recommendation. The experimental results on Netflix show that the new algorithm improves recommendation quality in MAE and coverage.