Dimension Reduction Based on Modified Maximum Margin Criterion for Tumor Classification

Based on Maximum margin criterion (MMC), a new algorithm, named modified MMC, is proposed for supervised dimensionality reduction in this paper. The algorithm aims at learning a linear transformation, and aims at maximizing the average margin between classes in the projected space. After projecting, the considered pair wise points within the same class are as close as possible, while those between different classes are as far as possible. The performance on two gene expression profiles datasets demonstrates the effectiveness of the proposed method.