Performance Analysis of Various Ensemble Feature Selection

  • Sandrilla R, Savitha Devi. M
Keywords: Ensemble Feature Selection (EFS), Boosting, Bagging, Stacking, Homogeneous, Heterogeneous


Machine learning modelling encourages the use of multiple models rather than a single model. In general, the most popular method of using multiple models is termed as ensemble where the individual model is grouped together. However, this method has a great impact on feature selection, in which the process of combining the model greatly enhances the better results. Special emphasis is given to elements that are relevant to initial idea. Feature selection is the process of locating and isolating important features while getting rid of outdated or irrelevant features in order to increase classifier accuracy. Hence the paper affords the detailed study for readers that helps them to better understand the basics and new approaches that are required to create an ensemble. We provide a summary of recent advances and predictions, just to lookout for changes in the future. Researchers will fail to discover new paths to excellence if they ignore all of the potential signals of performance in start-up review.  By this review the students are grasped to focus on a new subject intensely in order to reach their academic goals in ensemble learning. This study has been conducted to meet the new expectations with substantial number of publications. Once finished reading the content, one will be able to review it and search for easy ways, categories that are made up of distinct metrics. In order to better understand, recognise, and choose features, the work have created a detailed explanation of the feature selection approach. Real-world circumstances are perfect for EFS which may be accomplished using this technique

How to Cite
Savitha Devi. M, S. R. (2022). Performance Analysis of Various Ensemble Feature Selection. Design Engineering, (1), 228-247. Retrieved from