Using Artificial Neural Networks to Perform Feature Selection on Microarray Data
Giuliano Armano and Osvaldo Marullo
DMI, University of Cagliari, Italy
Abstract—This article illustrates a feature selection technique that makes use of artificial neural networks. The problem being faced is the analysis of microarray expression data, which requires a mandatory feature selection step due to the strong imbalance between number of features and size of the training set. The proposed technique has been assessed on relevant benchmark datasets. All datasets report gene expression levels taken from female subjects suffering from breast cancer against normal subjects. Experimental results, with average accuracy of about 84% and very good balance between specificity and sensitivity, point to the validity of the approach.
Copyright © 2022 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.
Index Terms—feature selection, artificial neural network, microarray data, cancer prediction
Cite: Giuliano Armano and Osvaldo Marullo, "Using Artificial Neural Networks to Perform Feature Selection on Microarray Data," International Journal of Pharma Medicine and Biological Sciences, Vol. 11, No. 3, pp. 54-58, July 2022. doi: 10.18178/ijpmbs.11.3.54-58Copyright © 2022 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.