| Nonnegative matrix factorization and its applications in pattern recognition |
| |
| 引用本文： | LlU Weixiang ZHENG Nanning YOU Qubo. Nonnegative matrix factorization and its applications in pattern recognition[J]. 科学通报(英文版), 2006, 51(1): 7-18. DOI: 10.1007/s11434-005-1109-6 |
| |
| 作者姓名： | LlU Weixiang ZHENG Nanning YOU Qubo |
| |
| 作者单位： | Institute of Artificial Intelligence and Robotics, Xi＇an Jiaotong University, Xi＇an 710049, China |
| |
| 基金项目： | Acknowledgments The authors would like to thank the anonymous reviewers for their constructive advice. This work was supported by the National Natural Science Foundation of China （Grant Nos. 60205001 and 60021302）. |
| |
| 摘 要： | Matrix factorization has been widely used in many fields[1]. In numerical algebra, the large-scale and com- plex problems can be transformed into small-scale sim- ple subproblems by matrix factorization; in applied statistics, low rank approximation of or…
|
| 关 键 词： | 特征抽取 NMF 模式识别 计算机 |
| 收稿时间： | 2005-08-13 |
| 修稿时间： | 2005-08-132005-11-02 |
Nonnegative matrix factorization and its applications in pattern recognition |
| |
| Weixiang Liu,Nanning Zheng,Qubo You. Nonnegative matrix factorization and its applications in pattern recognition[J]. Chinese science bulletin, 2006, 51(1): 7-18. DOI: 10.1007/s11434-005-1109-6 |
| |
| Authors: | Weixiang Liu Nanning Zheng Qubo You |
| |
| Affiliation: | (1) Institute of Artificial Intelligence and Robotics, Xi’an Jiaotong University, Xi’an, 710049, China |
| |
| Abstract: | Matrix factorization is an effective tool for large-scale data processing and analysis. Non- negative matrix factorization (NMF) method, which decomposes the nonnegative matrix into two non- negative factor matrices, provides a new way for ma- trix factorization. NMF is significant in intelligent information processing and pattern recognition. This paper firstly introduces the basic idea of NMF and some new relevant methods. Then we discuss the loss functions and relevant algorithms of NMF in the framework of probabilistic models based on our re- searches, and the relationship between NMF and information processing of perceptual process. Finally, we make use of NMF to deal with some practical questions of pattern recognition and point out some open problems for NMF. |
| |
| Keywords: | nonnegative data feature extraction NMF intrusion detection digital watermarking EEG signal analysis. |
| 本文献已被 维普 万方数据 SpringerLink 等数据库收录！ |
| 点击此处可从《科学通报(英文版)》浏览原始摘要信息 |
|
点击此处可从《科学通报(英文版)》下载全文 |
