Incremental nonnegative matrix factorization based on correlation and graph regularization for matrix completion
- North China Electric Power Univ., Beijing (China). School of Control and Computer Engineering; Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States). Computational Research Division
- North China Electric Power Univ., Beijing (China). Dept. of Mathematics and Physics
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States). Computational Research Division
Matrix factorization is widely used in recommendation systems, text mining, face recognition and computer vision. As one of the most popular methods, nonnegative matrix factorization and its incremental variants have attracted much attention. The existing incremental algorithms are established based on the assumption of samples are independent and only update the new latent variable of weighting coefficient matrix when the new sample comes, which may lead to inferior solutions. To address this issue, we investigate a novel incremental nonnegative matrix factorization algorithm based on correlation and graph regularizer (ICGNMF). The correlation is mainly used for finding out those correlated rows to be updated, that is, we assume that samples are dependent on each other. We derive the updating rules for ICGNMF by considering the correlation. We also present tests on widely used image datasets, and show ICGNMF reduces the error by comparing other methods.
- Research Organization:
- Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States). National Energy Research Scientific Computing Center (NERSC)
- Sponsoring Organization:
- USDOE
- OSTI ID:
- 1526987
- Journal Information:
- International Journal of Machine Learning and Cybernetics, Vol. 10, Issue 6; ISSN 1868-8071
- Country of Publication:
- United States
- Language:
- English
Similar Records
On the Equivalence of Nonnegative Matrix Factorization and K-means- Spectral Clustering
Incremental graph algorithms for parallel random access machines