基于NMFSC的特征基因提取

稀疏方法有一个重要的优点就是能减少基因表达数据的复杂度,故它具有很好的可解释性。在这篇文章中,我们利用稀疏控制的非负矩阵分解(NMFSC)来提取特征基因,因为稀疏控制的非负矩阵比其他稀疏方法更具有可解释性。在实验部分,将NMFSC应用在植物基因表达数据集上,并将其与传统的稀疏方法(SPCA)进行对比。实验证明我们的方法要比其他方法能提出更多的基因。

Feature Extraction for Genes Based on NMFSC

A significant advantage of sparse methods is that it can reduce the complicacy of genes expression data, which makes them easier good interpretability. In this article, we use non-negative matrix factorization with sparseness constraint （NMFSC） to extract features genes, because the non-negative matrix factorization with sparseness constraint is more explicable than that of other sparse methods. In the experimental part, NMFSC is used in plant gene expression data sets, and compared with the sparse traditional method （SPCA）. The experimental results show that our method can extract more feature genes than other methods.