Description

For removing irrelevant,redundant and noisy information from the data using feature selection algorithm.There are two approaches feature ranking, feature subset selection for feature selection.In this paper,propose new Feature ranking algorithm termed Mutiple Support Vector Machine-T-Statistics Feature Elimination Method for binary dataset. Feature selection with Feature ranking,used to identify the features from dataset. Dataset having hundreds and thousands of features.This is only for analyzing redundant features not be removed.MSVM-TRFE algorithms analyze the data for redundancies but may become computationally impractical on high-dimensional data sets.MSVM-TRFE algorithms analyze the data for redundancies but may become computationally impractical on high-dimensional data sets. And also By using MSVM-TRFE can generate subset selection and find ranking score to select a feature .MSVM-TRFE methods in the form of feature selection algorithm is used to get the perfect accuracy.