Title | ||
---|---|---|
Graph Regularized Non-negative Matrix with L0-Constraints for Selecting Characteristic Genes. |
Abstract | ||
---|---|---|
Non-negative Matrix Factorization (NMF) has been widely concerned in computer vision and data representation. However, the penalized and restriction L0-norm measure are imposed on the NMF model in traditional NMF methods. In this paper, we propose a novel graph regularized non-negative matrix with L0-constraints (GL0NMF) method which comprises the geometrical structure and a more interpretation sparseness measure. In order to extract the characteristic gene effectively, the steps are shown as follows. Firstly, the original data ( {mathbf{Q}} ) is decomposed into two non-negative matrices ( {mathbf{F}} ) and ( {mathbf{P}} ) by utilizing GL0NMF method. Secondly, characteristic genes are extracted by the sparse matrix ( {mathbf{F}} ). Finally, the extracted characteristic genes are validated by using Gene Ontology. In conclusion, the results demonstrate that our method can extract more genes than other conventional gene selection methods. |
Year | Venue | Field |
---|---|---|
2015 | ICIC | Adjacency matrix,Graph,Gene selection,Combinatorics,Pattern recognition,Matrix (mathematics),Matrix decomposition,Artificial intelligence,Degree matrix,Non-negative matrix factorization,Sparse matrix,Mathematics |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
13 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Chun-Xia Ma | 1 | 0 | 0.34 |
Gao Ying-Lian | 2 | 29 | 18.73 |
Dong Wang | 3 | 2 | 1.38 |
Jian Liu | 4 | 289 | 59.26 |
Liu Jin-Xing | 5 | 40 | 16.11 |