Feature Selection and Classification to Identify Cancer in Microarray Gene Expression Profile

Abstract

Cancer is one in all the dreadful diseases, which causes a substantial death rate in humans. Cancer is featured by associate irregular, unmanageable growth which will demolish and attack neighbouring healthy body tissues or somewhere else in the body. Microarray based mostly gene expression identification has been emerged as an economical technique for cancer classification, as well as for identification, prognosis, and treatment functions. In recent years, Deoxyribonucleic Acid microarray technique has gained a lot of attraction in both scientific and in industrial fields. It showed great importance in deciding the informative genes that can cause the cancer. This led to enhancements in early cancer diagnosis and in giving effective chemotherapy treatment. Studying cancer microarray gene expression data could be a difficult task because microarray is high dimensional-low sample dataset with loads of noisy or irrelevant genes and missing data. In this paper, we have a tendency to conduct a comprehensive study that focuses on exploring the main objectives and approaches that are applied using cancer microarray gene expression profile. We proceed by creating a classification for all approaches, and then conclude by investigating the foremost economical approaches that may be employed in this field.