The Optimization of Binary Feature Selection Problem using GSK Algorithm

In the data organizations, lots of tenders have produced datasets and decision-makers are required to apply several logical methods to extract valued data from databases, such as data removal, neural system, and statistical analysis. There has been a growth in real-world applications’ structures for machine learning and design acknowledgment.Among these features, an amount could be redundant or unrelated but it is hard to guess which one is valuable in the absence of previous skill. The algorithms for feature selection are commonly separated into three categories such as embedded, wrapper and filter. This research proposed the data imbalance and high dimensions for feature selection by using the GSK algorithm. The imbalanced data sets are categorized in two various types of approaches, namely, the wrapper and filter approach stage. The filter algorithm chooses topographies from the dataset and then customs the designated feature selection to teach the beginner. This research is focused to resolve the optimization of the binary feature selection problem. This comprehensive research supports the researchers in accomplishing a better solution for the feature selection optimization problem.