论文部分内容阅读
Background: Due to the development of the high throughput technologies, large bioinformatics data is produced.How to select the meaningful information from the big bioinformatics data becomes very crucial.Support vector machine-recursive feature elimination (SVM-RFE) is an efficient technique to filter out the discriminative information from the big data.According to the principle of SVM-RFE, the features having the smallest weight are eliminated iteratively.Since the bioinformatics data is usually very complex, some features may correlated to each other.The correlation among the features may affect the evaluation of some other features.