基于压缩感知的癌症基因表达数据分类

Classification of cancer gene expression data based on compressed sensing

提出了一种基于压缩感知原理的分类方法.把癌症基因表达数据分类问题归结为求解测试样本对于训练样本的稀疏表示问题,通过求解L1范数意义下的最优化问题来实现.提出的方法与Bagging神经网络和SVM的识别效果做了对比和分析,实验证明基于压缩感知的分类取得了相对较好的效果.

A classification method based on compressed sensing theory is proposed.The cancer gene expression classification problem was reduced to the problem as how to represent the testing samples from training data.The classification result thus could be achieved by solving the L1 norm-based optimization problem.We compared the effectiveness of this method with Bagging neutral network and SVM.Experiment results show that the compressed sensing-based classification method performs more effective.