Correntropy based feature selection using binary projection

Yan, Hui; Yuan, Xin; Yan, Shuicheng; Jing-yu, Yang

doi:10.1016/j.patcog.2011.04.014

Cited by 28 publications

(12 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tables 3 and 4 respectively detail the feature selection results of different algorithms on Final Training discretized by SNS and EF. 3,5,6,7,8,10,12,15,23,24,25,27, 29,31,32,36,37} ConsistencySubsetEval(BFBS) 15 {1, 3,6,8,12,13,24,31,[33][34][35][36][37]39,40} 2,3,5,6,9,12,13,20,23,24,27,28, 32,33,35,37,41} WrapperSubsetEval(BFBS) 21 {3, 5,7,8,9,10,13,14,15,17,18,19,21,23,25,26,27,30,32,35,37} 3,4,5,7,8,9,10,13,16,17,18,20,26 Especially, the classification accuracies of FSMRDE are always higher th...…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

A relative decision entropy-based feature selection approach

Jiang

Sui

Zhou

2015

Pattern Recognition

View full text Add to dashboard Cite

Section: Resultsmentioning

confidence: 99%

“…Feature selection can be understood in terms of Shannon's information theory [5]. Different feature selection methods based on information theoretic criteria have been proposed in recent years [6][7][8][9].…”

Section: Introductionmentioning

confidence: 99%

A relative decision entropy-based feature selection approach

Jiang

Sui

Zhou

2015

Pattern Recognition

View full text Add to dashboard Cite

“…Correntropy is a nonlinear similarity measure between two random variables, x and y [13,15,16], defined as …”

Section: Methodsmentioning

confidence: 99%

Non-negative matrix factorization by maximizing correntropy for cancer clustering

2013

View full text Add to dashboard Cite

BackgroundNon-negative matrix factorization (NMF) has been shown to be a powerful tool for clustering gene expression data, which are widely used to classify cancers. NMF aims to find two non-negative matrices whose product closely approximates the original matrix. Traditional NMF methods minimize either the l2 norm or the Kullback-Leibler distance between the product of the two matrices and the original matrix. Correntropy was recently shown to be an effective similarity measurement due to its stability to outliers or noise.ResultsWe propose a maximum correntropy criterion (MCC)-based NMF method (NMF-MCC) for gene expression data-based cancer clustering. Instead of minimizing the l2 norm or the Kullback-Leibler distance, NMF-MCC maximizes the correntropy between the product of the two matrices and the original matrix. The optimization problem can be solved by an expectation conditional maximization algorithm.ConclusionsExtensive experiments on six cancer benchmark sets demonstrate that the proposed method is significantly more accurate than the state-of-the-art methods in cancer clustering.

show abstract

“…The proof of Proposition 1 is based on convex conjugated function theory (Rockafellar 1997) and can be found in Yan et al (2011).…”

Section: Maximum Correntropy Criterionmentioning

confidence: 99%

FLAT-FIELDING OF SOLAR Hα OBSERVATIONS BASED ON THE MAXIMUM CORRENTROPY CRITERION

Zheng

et al. 2016

ApJ

View full text Add to dashboard Cite

The flat-field CCD calibration method of Kuhn et al. (KLL) is an efficient method for flat-fielding. However, since it depends on the minimum of the sum of squares error (SSE), its solution is sensitive to noise, especially nonGaussian noise. In this paper, a new algorithm is proposed to determine the flat field. The idea is to change the criterion of gain estimate from SSE to the maximum correntropy. The result of a test on simulated data demonstrates that our method has a higher accuracy and a faster convergence than KLL's and Chae's. It has been found that the method effectively suppresses noise, especially in the case of typical non-Gaussian noise. And the computing time of our algorithm is the shortest.

show abstract

Correntropy based feature selection using binary projection

Cited by 28 publications

References 39 publications

A relative decision entropy-based feature selection approach

A relative decision entropy-based feature selection approach

Non-negative matrix factorization by maximizing correntropy for cancer clustering

FLAT-FIELDING OF SOLAR Hα OBSERVATIONS BASED ON THE MAXIMUM CORRENTROPY CRITERION

Contact Info

Product

Resources

About