Review on feature selection methods in high dimensional domains

Feature selection has two important roles in the neuroimaging based classification. It possesses increased classification accuracy by eliminating the irrelevant features and identifying the best features for the discrimination of classes. Many approaches implemented for the feature selection in the context of neuroimaging. The development of feature selection methods is an active area of research. Now days a variety of feature selection methods are available so it is difficult to compare their characteristics To know this, different feature selection methods are compared and make a detailed study on this. This paper contains a comparison study among different feature selection methods and identification of merits and demerits.