Gene Selection Based on XGBoost and GWO for Microarray Data Classification

With the continuous development of DNA microarray technology, the increasing number of microarray data has entered a rapid development stage in the application of different fields. In recent years, cancer is the main cause of the shortening of the human life span. Using machine learning to process gene microarray datasets is very important for early cancer diagnosis. However, microarray datasets have the characteristics of small sample size, high dimensionality and uneven sample distribution. This paper uses the XGBoostGWO feature selection algorithm for cancer classification in microarray datasets. Firstly, XGBoost was used to sort genes to remove irrelevant genes. Finally, a grey wolf algorithm searches for the optimal gene subset from the gene subset obtained by the XGBoost. We conducted comprehensive experiments on 11 microarray datasets with two well-known learning classifiers. Compared with the most advanced feature selection methods, the experimental results show that the proposed method is superior to other feature selection methods in the evaluation indexes of accuracy, precision, F-score and recall.