A Hybrid Feature Selection Method Using Gene Expression Data

Feature Selection techniques are generally employed to remove the inessential attributes before machine learning technique could be applied. It thus plays an extremely important role by eliminating the unnecessary features that do not contribute and sometimes degrade the performance and prediction accuracy of the machine learning technique. With the growth of dimensionality of data, Feature Selection becomes even more important because it helps to reduce the dimensions of data and hence decreases the requisite memory and computational complexity of the machine learning techniques. Support vector machine-recursive feature elimination (SVM-RFE) has proven to be an efficient wrapper feature selection technique which continues to be widely utilized in many applications, especially in classification of gene expression data. From the perspective of this data, not only the precision in classification but also the stability of Feature Selection method plays an important role. Nonetheless, the topic of stability is ignored in study of feature selection algorithms. To improve the stability of RFE method, a fusion of Information Gain and RFE (IG-RFE-SVM) method is proposed in this paper. Experimental studies show that IG-RFE-SVM outperforms SVM-RFE method in terms of stability

Download Full-text

A Two-Stage Feature Selection Method for Gene Expression Data

OMICS A Journal of Integrative Biology ◽

10.1089/omi.2008.0083 ◽

2009 ◽

Vol 13 (2) ◽

pp. 127-137 ◽

Cited By ~ 15

Author(s):

Li-Yeh Chuang ◽

Chao-Hsuan Ke ◽

Hsueh-Wei Chang ◽

Cheng-Hong Yang

Keyword(s):

Gene Expression ◽

Feature Selection ◽

Gene Expression Data ◽

Feature Selection Method ◽

Selection Method ◽

Expression Data ◽

Two Stage

Download Full-text

An Integrated Feature Selection Algorithm for Cancer Classification using Gene Expression Data

Combinatorial Chemistry & High Throughput Screening ◽

10.2174/1386207322666181220124756 ◽

2019 ◽

Vol 21 (9) ◽

pp. 631-645 ◽

Cited By ~ 5

Author(s):

Saeed Ahmed ◽

Muhammad Kabir ◽

Zakir Ali ◽

Muhammad Arif ◽

Farman Ali ◽

...

Keyword(s):

Gene Expression ◽

Feature Selection ◽

Gene Expression Data ◽

Classification Accuracy ◽

Early Stage ◽

Small Sample Size ◽

Feature Selection Method ◽

Small Sample ◽

Expression Data ◽

Base Function

Aim and Objective: Cancer is a dangerous disease worldwide, caused by somatic mutations in the genome. Diagnosis of this deadly disease at an early stage is exceptionally new clinical application of microarray data. In DNA microarray technology, gene expression data have a high dimension with small sample size. Therefore, the development of efficient and robust feature selection methods is indispensable that identify a small set of genes to achieve better classification performance. Materials and Methods: In this study, we developed a hybrid feature selection method that integrates correlation-based feature selection (CFS) and Multi-Objective Evolutionary Algorithm (MOEA) approaches which select the highly informative genes. The hybrid model with Redial base function neural network (RBFNN) classifier has been evaluated on 11 benchmark gene expression datasets by employing a 10-fold cross-validation test. Results: The experimental results are compared with seven conventional-based feature selection and other methods in the literature, which shows that our approach owned the obvious merits in the aspect of classification accuracy ratio and some genes selected by extensive comparing with other methods. Conclusion: Our proposed CFS-MOEA algorithm attained up to 100% classification accuracy for six out of eleven datasets with a minimal sized predictive gene subset.

Download Full-text