This paper first analyzes the features of two classes of numerical methods for global analysis of nonlinear dynamical systems, which regard state space respectively as continuous and discrete ones. On basis of this un...This paper first analyzes the features of two classes of numerical methods for global analysis of nonlinear dynamical systems, which regard state space respectively as continuous and discrete ones. On basis of this understanding it then points out that the previously proposed method of point mapping under cell reference (PMUCR), has laid a frame work for the development of a two scaled numerical method suitable for the global analysis of high dimensional nonlinear systems, which may take the advantages of both classes of single scaled methods but will release the difficulties induced by the disadvantages of them. The basic ideas and main steps of implementation of the two scaled method, namely extended PMUCR, are elaborated. Finally, two examples are presented to demonstrate the capabilities of the proposed method.展开更多
为了将高维输入空间的数据映射到低维空间,利用可视化技术探测数据的固有特性,提出了用非线性主成分分析(NLPCA:NonLinear Principal Component Analysis)和自组织映射网络相结合的方法对生物信息学中基因表达数据进行聚类可视化分析。...为了将高维输入空间的数据映射到低维空间,利用可视化技术探测数据的固有特性,提出了用非线性主成分分析(NLPCA:NonLinear Principal Component Analysis)和自组织映射网络相结合的方法对生物信息学中基因表达数据进行聚类可视化分析。实验结果表明,该方法有较高的分类正确率,用于基因表达数据的聚类分析是行之有效的。展开更多
基金supported by the National Natural Science Foundation of China (NSFC) (10872155)
文摘This paper first analyzes the features of two classes of numerical methods for global analysis of nonlinear dynamical systems, which regard state space respectively as continuous and discrete ones. On basis of this understanding it then points out that the previously proposed method of point mapping under cell reference (PMUCR), has laid a frame work for the development of a two scaled numerical method suitable for the global analysis of high dimensional nonlinear systems, which may take the advantages of both classes of single scaled methods but will release the difficulties induced by the disadvantages of them. The basic ideas and main steps of implementation of the two scaled method, namely extended PMUCR, are elaborated. Finally, two examples are presented to demonstrate the capabilities of the proposed method.
文摘为了将高维输入空间的数据映射到低维空间,利用可视化技术探测数据的固有特性,提出了用非线性主成分分析(NLPCA:NonLinear Principal Component Analysis)和自组织映射网络相结合的方法对生物信息学中基因表达数据进行聚类可视化分析。实验结果表明,该方法有较高的分类正确率,用于基因表达数据的聚类分析是行之有效的。
