Robust graph regularized unsupervised feature selection. (15th April 2018)

• Record Type: Journal Article
• Authors: Tang, Chang; Zhu, Xinzhong; Chen, Jiajia; Wang, Pichao; Liu, Xinwang; Tian, Jie
• Is Part Of: Expert systems with applications. Volume 96(2018)
• Page Start: 64
• Page End: 76
• Publication Date: 2018-04-15
• Subjects: Unsupervised feature selection -- Local geometric structure -- Graph regularization -- Similarity preservation; 00-01 -- 99-00; Expert systems (Computer science) -- Periodicals; Systèmes experts (Informatique) -- Périodiques; Electronic journals; 006.33
• DOI: 10.1016/j.eswa.2017.11.053
• Languages: English
• ISSNs: 0957-4174

Abstract

Highlights: We propose a robust graph regularized unsupervised feature selection model. We introduce an l 1 -norm based graph to preserve the local structure of data. Feature reconstruction term is regularized by l 2, 1 -norm for robustness to outliers. An efficient solver is developed to solve the proposed optimization problem. Experiments confirmed our method outperforms other state-of-the-art methods. Abstract: Recent research indicates the critical importance of preserving local geometric structure of data in unsupervised feature selection (UFS), and the well studied graph Laplacian is usually deployed to capture this property. By using a squared l 2 -norm, we observe that conventional graph Laplacian is sensitive to noisy data, leading to unsatisfying data processing performance. To address this issue, we propose a unified UFS framework via feature self-representation and robust graph regularization, with the aim at reducing the sensitivity to outliers from the following two aspects: i) an l 2, 1 -norm is used to characterize the feature representation residual matrix; and ii) an l 1 -norm based graph Laplacian regularization term is adopted to preserve the local geometric structure of data. By this way, the proposed framework is able to reduce the effect of noisy data on feature selection. Furthermore, the proposed l 1 -norm based graph Laplacian is readily extendible, which can be easily integrated into other UFS methods and machine learning tasks with local geometrical structure of data being preserved. As demonstrated on ten challenging benchmark data sets, our algorithm significantly and consistently outperforms state-of-the-art UFS methods in the literature, suggesting the effectiveness of the proposed UFS framework.