Modified Neldermead Simplex Method for Clustering Gene Expression Data

In many numerical tests Nelder-Mead (NM) method succeeds in obtaining a good reduction in the function value using a relatively small number of function evaluations. Generally the microarray gene expression data dimension is high. However when optimizing high dimensional problems Nelder-Mead method suffers from poor convergence rate and early restart. To overcome this problem and to increase the global search area the Modified Nelder-Mead (MNM) is proposed. In the proposed work the expansion step is replaced by a new step called spread out. It is processed based on the assumption that a better point will be available apart from the best and good point. This may increase the global search considerably and will result in a better solution.