期刊
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
卷 32, 期 1, 页码 45-55出版社
IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2008.277
关键词
Nonnegative matrix factorization; singular value decomposition; clustering
资金
- University of Texas STARS Award
- US National Science Foundation (NSF) [DMS-0844497]
- NSF [IIS-0546280, DMS-0844513, 0509559]
- Microsoft Research
We present several new variations on the theme of nonnegative matrix factorization (NMF). Considering factorizations of the form X = FG(T), we focus on algorithms in which G is restricted to containing nonnegative entries, but allowing the data matrix X to have mixed signs, thus extending the applicable range of NMF methods. We also consider algorithms in which the basis vectors of F are constrained to be convex combinations of the data points. This is used for a kernel extension of NMF. We provide algorithms for computing these new factorizations and we provide supporting theoretical analysis. We also analyze the relationships between our algorithms and clustering algorithms, and consider the implications for sparseness of solutions. Finally, we present experimental results that explore the properties of these new methods.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据