Genetic data classification based on improved rank aggregation

The number of genes in microarray data is much larger than the number of available effective samples. Therefore, dealing with a small number of microarray data which are represented by high dimensional features, but of high correlations and strong interferences of redundancy and noise, has become one of the important tasks in gene microarray data extraction and classification. In this paper, a new method is proposed. We use wavelet decomposition to extract gene microarray data, and then use the t-test, ReliefF, Wilcoxon test, and other algorithms to select the data after wavelet transform. Finally, the Borda method is used to merge the sorted values. Three datasets were used in the experiment, namely the leukemia dataset, the prostate dataset, and the lung cancer dataset. Experimental results show that the method proposed in this paper can effectively classify the cancer gene microarray data.