Feature Selection in Big Data: Trends and Challenges

Abstract

Big data is a term used to represent data that is big in volume, speed, and variety. With inflammation, these characteristics are also inflated to 42 V’s. We have focused our survey for feature selection in big data, as feature selection is one of the most used methods for dimensionality reduction techniques. Feature selection is used for elimination of irrelevant and redundant features from dataset to improve the classification performance. This paper includes big data characteristics, different feature selection method, and current research challenges of feature selection. We observed that swarm intelligence techniques are the most popular methods among researchers for feature selection in big data. Further, we conclude that gray wolf optimization and particle swarm optimization are the most preferred algorithms by researchers.