针对基于稀疏回归的多标签特征选择方法中数据的特征和标签之间线性关系假设不成立的问题,提出一种基于依赖最大化和稀疏回归的多标签特征选择方法(multi-label feature selection with dependence maximization and sparse regression,...针对基于稀疏回归的多标签特征选择方法中数据的特征和标签之间线性关系假设不成立的问题,提出一种基于依赖最大化和稀疏回归的多标签特征选择方法(multi-label feature selection with dependence maximization and sparse regression,DMSR)。构建数据的低维子空间,最大化低维空间与数据的标签空间之间的依赖性,使用希尔伯特-施密特独立性准则作为依赖性的计算依据,将数据从特征空间映射到该低维空间,设计一种交替优化的算法对稀疏回归模型进行求解,得到用于特征选择的投影矩阵。在多个不同类型的多标签数据集上的实验结果表明,所提算法的性能优于其它对比算法。展开更多
In order to overcome the shortcomings that the reconstructed spectral reflectance may be negative when using the classic principal component analysis （PCA）to reduce the dimensions of the multi-spectral data, a nonne...In order to overcome the shortcomings that the reconstructed spectral reflectance may be negative when using the classic principal component analysis （PCA）to reduce the dimensions of the multi-spectral data, a nonnegative constrained principal component analysis method is proposed to construct a low-dimensional multi-spectral space and accomplish the conversion between the new constructed space and the multispectral space. First, the reason behind the negative data is analyzed and a nonnegative constraint is imposed on the classic PCA. Then a set of nonnegative linear independence weight vectors of principal components is obtained, by which a lowdimensional space is constructed. Finally, a nonlinear optimization technique is used to determine the projection vectors of the high-dimensional multi-spectral data in the constructed space. Experimental results show that the proposed method can keep the reconstructed spectral data in [ 0, 1 ]. The precision of the space created by the proposed method is equivalent to or even higher than that by the PCA.展开更多
文摘针对基于稀疏回归的多标签特征选择方法中数据的特征和标签之间线性关系假设不成立的问题,提出一种基于依赖最大化和稀疏回归的多标签特征选择方法(multi-label feature selection with dependence maximization and sparse regression,DMSR)。构建数据的低维子空间,最大化低维空间与数据的标签空间之间的依赖性,使用希尔伯特-施密特独立性准则作为依赖性的计算依据,将数据从特征空间映射到该低维空间,设计一种交替优化的算法对稀疏回归模型进行求解,得到用于特征选择的投影矩阵。在多个不同类型的多标签数据集上的实验结果表明,所提算法的性能优于其它对比算法。
基金The Pre-Research Foundation of National Ministries andCommissions (No9140A16050109DZ01)the Scientific Research Program of the Education Department of Shanxi Province (No09JK701)
文摘In order to overcome the shortcomings that the reconstructed spectral reflectance may be negative when using the classic principal component analysis （PCA）to reduce the dimensions of the multi-spectral data, a nonnegative constrained principal component analysis method is proposed to construct a low-dimensional multi-spectral space and accomplish the conversion between the new constructed space and the multispectral space. First, the reason behind the negative data is analyzed and a nonnegative constraint is imposed on the classic PCA. Then a set of nonnegative linear independence weight vectors of principal components is obtained, by which a lowdimensional space is constructed. Finally, a nonlinear optimization technique is used to determine the projection vectors of the high-dimensional multi-spectral data in the constructed space. Experimental results show that the proposed method can keep the reconstructed spectral data in [ 0, 1 ]. The precision of the space created by the proposed method is equivalent to or even higher than that by the PCA.
