FeatureSelect

FeatureSelect is a software tool for feature selection, an essential preprocessing step in various scientific fields. Unlike existing tools like WEKA, which primarily use filter methods, FeatureSelect incorporates both filter methods and wrapper methods, including optimization algorithms and three types of learners. This software provides a user-friendly interface for feature selection and can be applied to balanced and unbalanced datasets using various score functions such as accuracy, sensitivity, and specificity.

FeatureSelect implements ten efficient and well-known optimization algorithms, including the recently developed Wrapper Classifier Corrector (WCC) algorithm. The software was tested on different datasets, and the results showed that the algorithms' performance varies depending on the dataset.