UNSTRUCTURED
Alzheimer’s disease (AD) is a progressive neurodegenerative brain disorder with complex genetic architecture. This disease is the focus point of many bioinformatics kinds of research, where the key goal of these researches is to classify the genes involved in the processes of Alzheimer's and to explore the function of these risk genes in the progress of the disease. For this purpose, we here seek out the best model to detect the genes related to AD with the usage of several feature selection methods. In this study, we compared the efficiency of the feature selection methods with SVM classifier including mRMR, CFS, Chi-Square Test, F-score, and GA was compared. The accuracy of SVM classifier has been calculated with validation methods such as 10-fold cross-validation. We applied these methods to the public AD gene expression dataset consist of 696 samples and 200 gene. The results show that mRMR and F-score obtain high accuracy around 84% with number of genes between 40 to 80.