RENT—Repeated Elastic Net Technique for Feature Selection

Feature selection is an essential step in data science pipelines to reduce the complexity associated with large datasets. While much research on this topic focuses optimizing predictive performance, few studies investigate stability context of feature process. In study, we present Repeated Elastic Net Technique (RENT) for Selection. RENT uses ensemble generalized linear models elastic net regularization, each trained distinct subsets training data. The based three criteria evaluating weight distributions features across all elementary models. This fact leads high that improve robustness final model. Furthermore, unlike established selectors, provides valuable information model interpretation concerning identification objects are difficult predict during training. our experiments, benchmark against six selectors eight multivariate datasets binary classification and regression. experimental comparison, shows a well-balanced trade-off between performance stability. Finally, underline additional interpretational value exploratory post-hoc analysis healthcare dataset.