核心概念
A novel method, GRROOR, optimizes feature selection in multi-label classification using orthogonal regression.
要約
The article introduces GRROOR, a method for multi-label feature selection using orthogonal regression. It addresses the challenge of preserving discriminative information in multi-label data by considering global redundancy and relevance. The method optimizes feature selection by incorporating orthogonal regression with feature weighting. Extensive experiments on ten datasets demonstrate the effectiveness of GRROOR.
統計
Least square regression-based multi-label feature selection methods learn a projection matrix W with sparsity restriction by minimizing regression error.
The proposed GRROOR method introduces global redundancy and relevance information into the orthogonal regression model.
引用
"The method employs orthogonal regression with feature weighting to retain sufficient statistical and structural information related to local label correlations."