Robust inner product regularized unsupervised feature selection

Qian, Youcheng; Yin, Xueyan; Gao, Wei

doi:10.1007/s11042-019-08159-y

Cited by 17 publications

(26 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Many unsupervised feature selection methods, both similarity preserving (filter) [9,19] and embedded [6,8,10,14] methods, are based on input data alone and rarely take the advantage of the external sources of knowledge about feature group structures. The feature groups used by some feature selection methods are also formed with input data [15,18].…”

Section: Related Workmentioning

confidence: 99%

“…Number of groups is set to 0.04 of the original feature set based on the previous findings for MT dataset [3]. Baselines: As baselines, we use LS algorithm and Spectral Feature Selection SPEC [19] as similarity preserving methods and Multi Cluster Feature Selection (MCFS) [6], Robust Unsupervised Feature Selection (RUFS) [14] and HUFS as embedded methods. RUFS has proven high performance compared to many existing embedded methods and HUFS uses feature group information similar to our method.…”

Section: Feature Groupingmentioning

confidence: 99%

“…Many unsupervised feature selection methods evaluate features using instance-feature data alone, which is available in the form of the data matrix [9,14]. In contrast, recent work shows that features can be grouped according to various criteria and this group information can improve the usefulness of the feature selection [17].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Group Based Unsupervised Feature Selection

Perera

Chan

Karunasekera

2020

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

Unsupervised feature selection is an important task in machine learning applications, yet challenging due to the unavailability of class labels. Although a few unsupervised methods take advantage of external sources of correlations within feature groups in feature selection, they are limited to genomic data, and suffer poor accuracy because they ignore input data or encourage features from the same group. We propose a framework which facilitates unsupervised filter feature selection methods to exploit input data and feature group information simultaneously, encouraging features from different groups. We use this framework to incorporate feature group information into Laplace Score algorithm. Our method achieves high accuracy compared to other popular unsupervised feature selection methods (∼30% maximum improvement of Normalized Mutual Information (NMI)) with low computational costs (∼50 times lower than embedded methods on average). It has many real world applications, particularly the ones that use image, text and genomic data, whose features demonstrate strong group structures.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Feature Groupingmentioning

confidence: 99%

See 1 more Smart Citation

Group Based Unsupervised Feature Selection

Perera

Chan

Karunasekera

2020

Advances in Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

show abstract

“…The use of the 2, 1-norm penalty on a coefficient matrix is convenient for feature selection, as it promotes row sparsity, and can be found in several other works [13], [14], [15], [16], [17], [18]. Some recent works propose the use of a regularized coefficient matrix to choose features that can reconstruct the entire data set well in a manner that is similar to the problem we tackle in this paper [18], [19].…”

Section: Locality and Cluster Structure Preservationmentioning

confidence: 99%

Regularized greedy column subset selection

Ordozgoiti

Mozó

Lacalle

2019

Information Sciences

View full text Add to dashboard Cite

The Column Subset Selection Problem provides a natural framework for unsupervised feature selection. Despite being a hard combinatorial optimization problem, there exist efficient algorithms that provide good approximations. The drawback of the problem formulation is that it incorporates no form of regularization, and is therefore very sensitive to noise when presented with scarce data. In this paper we propose a regularized formulation of this problem, and derive a correct greedy algorithm that is similar in efficiency to existing greedy methods for the unregularized problem. We study its adequacy for feature selection and propose suitable formulations. Additionally, we derive a lower bound for the error of the proposed problems. Through various numerical experiments on real and synthetic data, we demonstrate the significantly increased robustness and stability of our method, as well as the improved conditioning of its output, all while remaining efficient for practical use.

show abstract

“…However, such heuristic based methods usually ignore the correlation among the features and redundancy may exist in the selected features. In recent years, different methods [26] [12] [24] [23] have been proposed to evaluate feature quality jointly. Linear projection based methods [26] [8] [4] [19] with sparsity-inducing L 2,1 norm have become prevalent among others.…”

Section: Related Workmentioning

confidence: 99%

Multi-view unsupervised feature selection by cross-diffused matrix alignment

Wei

Cao

2017

2017 International Joint Conference on Neural Networks (IJCNN)

View full text Add to dashboard Cite

Multi-view high-dimensional data become increasingly popular in the big data era. Feature selection is a useful technique for alleviating the curse of dimensionality in multi-view learning. In this paper, we study unsupervised feature selection for multi-view data, as class labels are usually expensive to obtain. Traditional feature selection methods are mostly designed for single-view data and cannot fully exploit the rich information from multi-view data. Existing multi-view feature selection methods are usually based on noisy cluster labels which might not preserve sufficient information from multi-view data. To better utilize multi-view information, we propose a method, CDMA-FS, to select features for each view by performing alignment on a cross diffused matrix. We formulate it as a constrained optimization problem and solve it using Quasi-Newton based method. Experiments results on four real-world datasets show that the proposed method is more effective than the state-of-theart methods in multi-view setting.

show abstract

Robust inner product regularized unsupervised feature selection

Cited by 17 publications

References 41 publications

Group Based Unsupervised Feature Selection

Group Based Unsupervised Feature Selection

Regularized greedy column subset selection

Multi-view unsupervised feature selection by cross-diffused matrix alignment

Contact Info

Product

Resources

About