Mechanical Engineering, K. N. Toosi University of Technology, Tehran, Iran
Abstract
The small number of samples and inherent variability in laboratory and biological processes hinder the analysis of high-volume microarray data. The first issue arises from the increase in computational cost and complexity of the classifications. The second issue involves the reduction in the classification's ability to generalize and predict new samples, which diminishes its validity. Thirdly, the high number of genes used as features compared to sample size in certain data sets increases the likelihood of inappropriate genes being used for gene classification in predictive models. Furthermore, interpreting disease-causing genes is complicated because only a small subset of genes can provide a more precise biological description of the disease. Therefore, a smaller set of gene expression data should be focused on in order to achieve a more effective elucidation of genes containing information. Thus, in microarray data analysis, the main objective is to significantly reduce the number of genes by selecting discriminative genes during the classification process, a measure known as gene selection. This article employs gene expression classification on colon cancer, breast cancer, leukemia, prostate tumours, and DLBCL, and each is evaluated independently in the feature selection cycle and classified using a varying number of features. This article employs gene expression classification on colon cancer, breast cancer, leukemia, prostate tumours, and DLBCL, and each is evaluated independently in the feature selection cycle and classified using a varying number of features.
Turki, F., Khadem, A., & Aghchei, A. (2023). Investigation and Analysis of Gene Expression Using the Fusion Method of Feature Selection and Dynamic Neural Network Classification. Transactions on Machine Intelligence, 6(2), 11-20.
MLA
F. Turki; A. Khadem; A.H. Jalali Aghchei. "Investigation and Analysis of Gene Expression Using the Fusion Method of Feature Selection and Dynamic Neural Network Classification". Transactions on Machine Intelligence, 6, 2, 2023, 11-20.
HARVARD
Turki, F., Khadem, A., Aghchei, A. (2023). 'Investigation and Analysis of Gene Expression Using the Fusion Method of Feature Selection and Dynamic Neural Network Classification', Transactions on Machine Intelligence, 6(2), pp. 11-20.
VANCOUVER
Turki, F., Khadem, A., Aghchei, A. Investigation and Analysis of Gene Expression Using the Fusion Method of Feature Selection and Dynamic Neural Network Classification. Transactions on Machine Intelligence, 2023; 6(2): 11-20.
)