A parsimonious threshold-independent protein feature selection method through the area under receiver operating characteristic curve

Wang, Zhanfeng; Chang, Yuan‐chin Ivan; Ying, Zhiliang; Zhu, Liang; Yang, Yaning

doi:10.1093/bioinformatics/btm442

Cited by 41 publications

(44 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Thus, we follow Ma and Huang [30] and Wang et al [31], to use a sigmoid function S(t)=1/(1+exp(-t)) to approximate ψ(•). Consequently, the smoothed AUC estimate is defined as,…”

Section: Linear Combination Of Genes Within a Gene Setmentioning

confidence: 99%

See 1 more Smart Citation

Identifying Differential Gene Sets using the Linear Combination of Genes with Maximum AUC

Wang¹,

Tsai²,

Chang³

2012

jpb

View full text Add to dashboard Cite

“…Thus, we follow Ma and Huang [30] and Wang et al [31], to use a sigmoid function S(t)=1/(1+exp(-t)) to approximate ψ(•). Consequently, the smoothed AUC estimate is defined as,…”

Section: Linear Combination Of Genes Within a Gene Setmentioning

confidence: 99%

“…A discussion of the proposed approach is presented in Section 4. For other properties of AUC and its applications, please refer to Metz et al [25], Su and Liu [26], Zhou et al [27], Pepe [28], Liu et al [29], Ma and Huang [30] and Wang et al [31], and the references therein.…”

Section: Introductionmentioning

confidence: 99%

Identifying Differential Gene Sets using the Linear Combination of Genes with Maximum AUC

Wang¹,

Tsai²,

Chang³

2012

jpb

View full text Add to dashboard Cite

“…In order to facilitate the maximization process, the standard normal distribution function is used in place of H(z) [9], or a sigmoid approximation for this purpose is also proposed in [11] and [12]. Here, we consider the former approximation:…”

Section: Area Under the Roc Curvementioning

confidence: 99%

Boosting Learning Algorithm for Pattern Recognition and Beyond

Komori

Eguchi

2011

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYThis paper discusses recent developments for pattern recognition focusing on boosting approach in machine learning. The statistical properties such as Bayes risk consistency for several loss functions are discussed in a probabilistic framework. There are a number of loss functions proposed for different purposes and targets. A unified derivation is given by a generator function U which naturally defines entropy, divergence and loss function. The class of U-loss functions associates with the boosting learning algorithms for the loss minimization, which includes AdaBoost and LogitBoost as a twin generated from Kullback-Leibler divergence, and the (partial) area under the ROC curve. We expand boosting to unsupervised learning, typically density estimation employing U-loss function. Finally, a future perspective in machine learning is discussed.

show abstract

“…; 유전자 발현 데 이터 분석 [11,12,15]; DNA 메틸화 데이터 분석 [18,27]; 단백질 데이터 분석 [24,26] 해 그동안의 기법들은 t-test [16], wilcoxon ranksum test [22]와 같이 각 변수들을 한 개씩 독립적 으로 선택하는 것이 대부분이었다 [21]. 그 외에는 동시에 여러 개의 변수를 고려하는 방법으로 SVM 을 이용한 연구와(예 : [13,25] 의 알고리즘을 사용하였다.…”

Section: Introductionunclassified

Set Covering-based Feature Selection of Large-scale Omics Data

Ma¹,

Yan²,

Kim³

et al. 2014

Journal of the Korean Operations Research and Management Scienc

View full text Add to dashboard Cite

In this paper, we dealt with feature selection problem of large-scale and high-dimensional biological data such as omics data. For this problem, most of the previous approaches used simple score function to reduce the number of original variables and selected features from the small number of remained variables. In the case of methods that do not rely on filtering techniques, they do not consider the interactions between the variables, or generate approximate solutions to the simplified problem. Unlike them, by combining set covering and clustering techniques, we developed a new method that could deal with total number of variables and consider the combinatorial effects of variables for selecting good features. To demonstrate the efficacy and effectiveness of the method, we downloaded gene expression datasets from TCGA (The Cancer Genome Atlas) and compared our method with other algorithms including WEKA embeded feature selection algorithms. In the experimental results, we showed that our method could select high quality features for constructing more accurate classifiers than other feature selection algorithms.

show abstract

A parsimonious threshold-independent protein feature selection method through the area under receiver operating characteristic curve

Abstract: Supplementary data are available at Bioinformatics online.

Cited by 41 publications

References 12 publications

Identifying Differential Gene Sets using the Linear Combination of Genes with Maximum AUC

Identifying Differential Gene Sets using the Linear Combination of Genes with Maximum AUC

Boosting Learning Algorithm for Pattern Recognition and Beyond

Set Covering-based Feature Selection of Large-scale Omics Data

Contact Info

Product

Resources

About