Symmetrical uncertainty method to extract essential features for Endoscopic Gastritis data set

Abstract

Classifying high dimensional numerical data is an exceptionally difficult issue. High dimensional data for example data sets with hundreds or thousands of features, can contain high degree of irrelevant and redundant information which greatly degrades the performance of learning algorithms. Therefore, feature selection becomes necessary for machine learning tasks for facing high dimensional data. To address this issue, an efficient feature selection method using symmetrical uncertainty is used to facilitate classifying high-dimensional numerical data. The focus here is on feature selection method that are able to assess the goodness or ranking of the individual features. The threshold method used here helps to accurately determine which features is relevant and which features is redundant. The relevant features is called as the essential features while the irrelevant features will be ignore from feature classification.