面对模式分析、数据挖掘中海量数据,降维算法已经成为科学研究人员最为

强有力的工具.对降维算法的研究具有很高的学术价值和应用潜力.本文较为详

细的回顾了现有的降维算法,以及他们在模式分析中的应用.在此基础上,着眼于

提高嵌入空间的不同类别的样本之间的距离,我们提出了两种有监督情形下的流

形学习算法.模拟和实际数据都显示了有监督流形学习算法的良好的性能.-Face pattern analysis, data mining massive data, dimension reduction algorithm has become the most powerful scientific tool for staff. Dimensionality reduction algorithm of high academic value and potential applications of this paper a more detailed review of the existing dimensionality reduction algorithms and their applications in pattern analysis. On this basis, focusing on improving the embedding space, different types of distance between samples, we proposed two kinds of cases supervised manifold learning algorithm simulation and actual data have shown a supervised manifold learning algorithm good performance.