Combined mRMR filter and sparse Bayesian classifier for analysis of gene expression data

Soltani, Mehran; Shammakhi, Mohammad Hasan; Khorram, Saeed; Sheikhzadeh, Hamid

doi:10.1109/icspis.2016.7869891

Cited by 3 publications

(5 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The parallelized mRMR method helps in shortens the computation time compared to the conventional single method. Therefore, the maximum relevance and minimum redundancy (mRMR) feature selection algorithm and parallelized mRMR method increase the accuracy if the smaller subset of genes selection for cancer classification in this research [28], [29].…”

Section: Discussionmentioning

confidence: 99%

An Improved Parallelized mRMR for Gene Subset Selection in Cancer Classification

Kusairi

Moorthy

Haron

et al. 2017

International Journal on Advanced Science, Engineering and Information Technology

View full text Add to dashboard Cite

Abstract-DNA microarray technology has become a more attractive tool for cancer classification in the scientific and industrial fields. Based on the previous researchers, the conventional approach for cancer classification is primarily based on the morphological appearance of the tumor. The limitations of this approach are the bias in identify the tumors by expert and faced the difficulty in differentiating the cancer subtypes due to most cancers being highly related to the specific biological insight. Thus, this study proposes an improved parallelized Minimum Redundancy Maximum Relevance (mRMR), which is a particularly fast feature selection method for finding a set of both relevant and complementary features. The mRMR can identify genes more relevance to the biological context that leads to richer biological interpretations. The proposed method is expected to achieve accurate classification performance using a small number of predictive genes when tested using two datasets from Cancer Genome Project and compared to previous methods.

show abstract

Section: Discussionmentioning

confidence: 99%

An Improved Parallelized mRMR for Gene Subset Selection in Cancer Classification

Kusairi

Moorthy

Haron

et al. 2017

International Journal on Advanced Science, Engineering and Information Technology

View full text Add to dashboard Cite

show abstract

“…Wrapper methods search for features according to a specific learning algorithm. Finally, embedded methods [11] selects features in the learning process [2].…”

Section: Related Work (Mrmr Feature Selection)mentioning

confidence: 99%

“…In feature selection algorithms based on information, the main goal is to find a feature set S that has the greatest dependence on the target class z [2]. Using the maximal-dependency criterion, the mutual information relationship is defined as follows:…”

Section: Max-relevance and Min-redundancymentioning

confidence: 99%

See 1 more Smart Citation

Feature Selection Method Based on Differential Correlation Information Entropy

Wang

Yan

2020

Neural Process Lett

View full text Add to dashboard Cite

Feature selection is one of the major aspects of pattern classification systems. In previous studies, Ding and Peng recognized the importance of feature selection and proposed a minimum redundancy feature selection method to minimize redundant features for sequential selection in microarray gene expression data. However, since the minimum redundancy feature selection method is used mainly to measure the dependency between random variables of mutual information, the results cannot be optimal without consideration of global feature selection. Therefore, based on the framework of minimum redundancy-maximum correlation, this paper introduces entropy to measure global feature selection and proposes a new feature subset evaluation method, differential correlation information entropy. In our function, different bivariate correlation metrics are selected. Then, the feature selection is completed through sequence forward search. Two different classification models are used on eleven standard data sets of the UCI machine learning knowledge base to compare various comparison algorithms, such as mRMR, reliefF and feature selection method with joint maximal information entropy, with our method. The experimental results show that feature selection based on our proposed method is obviously superior to that of other models.

show abstract

“…Multivariate methods pose greater complexity than univariate ones in that, besides requiring a method to evaluate groups of features, they also involve a search mechanism in the space of all possible feature subsets. Regarding the first issue, many works employ correlationoriented criteria based on the concept of mutual information (MI) (see, e.g., [12], [13], [8], [14], [15], [16]). Indeed, the MI between the features and the output reveals their discriminating capabilities, whereas the correlation among features indicates possible redundancy issues (see, e.g., the Minimal Redundancy Maximal Relevance (MRMR) algorithm [14], and the Correlation Based Filtering (CFS) method [17]).…”

Section: Introductionmentioning

confidence: 99%

A Distributed Feature Selection Algorithm Based on Distance Correlation with an Application to Microarrays

Brankovic

Hosseini

Piroddi

2019

IEEE/ACM Trans. Comput. Biol. and Bioinf.

View full text Add to dashboard Cite

DNA microarray datasets are characterized by a large number of features with very few samples, which is a typical cause of overfitting and poor generalization in the classification task. Here we introduce a novel feature selection (FS) approach which employs the distance correlation (dCor) as a criterion for evaluating the dependence of the class on a given feature subset. The dCor index provides a reliable dependence measure among random vectors of arbitrary dimension, without any assumption on their distribution. Moreover, it is sensitive to the presence of redundant terms. The proposed FS method is based on a probabilistic representation of the feature subset model, which is progressively refined by a repeated process of model extraction and evaluation. A key element of the approach is a distributed optimization scheme based on a vertical partitioning of the dataset, which alleviates the negative effects of its unbalanced dimensions. The proposed method has been tested on several microarray datasets, resulting in quite compact and accurate models obtained at a reasonable computational cost.

show abstract

Combined mRMR filter and sparse Bayesian classifier for analysis of gene expression data

Cited by 3 publications

References 12 publications

An Improved Parallelized mRMR for Gene Subset Selection in Cancer Classification

An Improved Parallelized mRMR for Gene Subset Selection in Cancer Classification

Feature Selection Method Based on Differential Correlation Information Entropy

A Distributed Feature Selection Algorithm Based on Distance Correlation with an Application to Microarrays

Contact Info

Product

Resources

About