Authors : R. Mythily and W. Aisha Banu
Abstract: Feature selection and feature extraction is a preprocessing step for the effective analysis of high dimensional data. They widely used nowadays to improve the performance such as estimated accuracy, low computational performance and reducing storage. Optimization algorithms give better results but the process involved to find the optimal features are expensive. A lot of data are generating from various sources like the internet, weather forecast, medical data, etc. with huge in dimensions along with noise. The main objective is to extract useful information by reducing the dimensions of the data and to increase the efficiency and accuracy of the result. Feature selection is one of the methods used in dimensionality reduction the main aim is to select the small subset of best features from the original feature set without affecting the originality of data. Analyzing the suitable feature selection method for high dimensional data is very essential and hence it is required to survey on the various feature selection methods to improve the performance of the machine learning task. This study covers the concept of feature selection and its various methods and algorithms. We conclude this research with future research directions.
R. Mythily and W. Aisha Banu, 2017. Feature Selection for Optimization Algorithms: Literature Survey. Journal of Engineering and Applied Sciences, 12: 5735-5739.