Feature Selection of Microarray Data Using Simulated Kalman Filter with Mutation

Microarrays have been proven to be beneficial for understanding the genetics of disease. They are used assess many different types cancers. Machine learning algorithms, like artificial neural network (ANN), can trained determine whether a microarray sample is cancerous or not. The classification performed using features DNA data, which composed thousands gene values. However, most values uninformative and redundant. Meanwhile, number samples significantly smaller in comparison genes. Therefore, this paper proposed use simulated Kalman filter with mutation (SKF-MUT) feature selection data enhance accuracy ANN. algorithm based on metaheuristics optimization algorithm, inspired by famous estimator. operator performance original SKF features. Eight benchmark datasets were used, comprised: diffuse large b-cell lymphomas (DLBCL); prostate cancer; lung leukemia “small, round blue cell tumor” (SRBCT); brain tumor; nine human tumors; 11 tumors. These consist both binary multiclass datasets. taken as measurement considering confusion matrix. Based results, SKF-MUT effectively selected needed, leading toward higher ranging from 95% 100%.