机构地区: 国防科学技术大学机电工程与自动化学院
出 处: 《系统仿真学报》 2007年第15期3380-3382,3386,共4页
摘 要: 肿瘤表达谱分类是一个典型的高纬度小样本分类问题。在测试的基因表达谱中存在大量的非差异表达冗余基因,通过一个有效的基因预选择策略得到一个较小的候选基因子集,然后利用免疫优化算法对最优基因子集和分类器参数进行优化,从而得到一个更好的分类预测模型。在四个真实的肿瘤表达谱数据上,与几种不同的方法进行了比较,结果显示新方法可以得到更好的分类精度,同时表现出很好的稳定性。 Cancer microarray expression classification is a typical case that has high dimensions and small samples. So the most relevant gene selection and parameter optimization for classifier are both important issues. A robust two-step approach was proposed. First a pre-possessed strategy selected a relevant candidate gene subset; secondly, an immune clonal algorithm simultaneously optimized the gene subset and parameters for support vector machine. On four real cancer microarray datasets, the new approach was compared to the several existing methods. The experimental results show that the proposed approach can achieve high classification accuracy and is more robust.
领 域: [自动化与计算机技术] [自动化与计算机技术]