Supervised Model for Brain Stroke Using Feature Selection Approaches

Feature selection has been considered as a vital role in machine learning (ML) for various data analysis on classification or the importance of features for classification. Generally, ML employs multiple diverse approaches to solve different problems as needed by the experimental demand. Since feature selection is part of ML, many complex problems can be solved using feature selection techniques. Feature selection can solve the problem by selecting features from the dataset's own pool. The main aim of this feature selection approach is to select the practical features from the dataset to reduce dimensionality with effective accuracy performance. Thus, supervised model-based feature selection approaches are considered for analyzing healthcare datasets to predict diseases. Several methods for supervised models are considered to select features from the brain stroke dataset for predicting disease, based on various experiments. The proposed strategies are demonstrated using machine learning methods for various experimental performances.