Feature subset selection Filter–Wrapper based on low quality data

Highlights：

• Low quality data (LQD) appear in real problems. We incorporate the processing of LQD in feature selection.

• We propose a feature selection method that works within the framework of the fuzzy logic theory.

• Our approach is classified as a hybrid method that combines the filter and wrapper methods.

• The experiments evaluate the performance of the approach with low quality and microarray datasets.

• The results indicate that the approach has a good classification performance with or without LQDs.