Chen_et_al_2022.pdf (1.55 MB)
Graph non-negative matrix factorization with alternative smoothed L0 regularizations
journal contribution
posted on 2023-08-30, 20:01 authored by Keyi Chen, Hangjun Che, Xinqi Li, Man-Fai LeungGraph non-negative matrix factorization (GNMF) can discover the data’s intrinsic low-dimensional structure embedded in the high-dimensional space. So, it has superior performance for data representation and clustering. Unfortunately, it is sensitive to noise and outliers. In this paper, to improve the robustness of GNMF, l0 norm is introduced to enhance the sparsity of factorized matrices. As the discontinuity of l0 norm and minimizing it is a NP-hard problem, five functions approximating l0 norm are used to transform the problem of the sparse graph non-negative matrix factorization (SGNMF) to a global optimization problem. Finally, the multiplicative updating rules (MUR) are designed to solve the problem and the convergence of algorithm is proven. In the experiment, the accuracy and normalized mutual information of clustering results show the superior performance of SGNMF on five public datasets.
History
Refereed
- Yes
Volume
0Issue number
0Page range
0Publication title
Neural Computing and ApplicationsISSN
1433-3058External DOI
Publisher
SpringerFile version
- Accepted version
Language
- eng
Official URL
Legacy posted date
2022-06-15Legacy creation date
2022-06-15Legacy Faculty/School/Department
Faculty of Science & EngineeringUsage metrics
Categories
No categories selectedKeywords
Licence
Exports
RefWorks
BibTeX
Ref. manager
Endnote
DataCite
NLM
DC