Feature Selection using MultiObjective Grey Wolf Optimization Algorithm

Deepak Gupta, Nimish Verma, Mayank Sehgal, Nitesh
2019 Zenodo  
Multi Objective Grey wolf Optimization is one a meta-heuristic technique. The MOGWO has recently gained a huge research interest from numerous domains due to its impressive characteristics over other meta-heuristics optimization techniques: it has less parameters, derivation information is not required in the initial stage, scalable, flexible, easy to use. In this paper MOGWO, which is based on the leadership hunting technique of grey wolves is used for feature selection. The traditional GWO is
more » ... useful for single objective optimization problems. Since, feature extraction is a multi-objective problem; this paper utilizes multiobjective GWO algorithm. In this paper, MOGWO is applied to 6 different datasets to understand its application in diverse set of problems. At first, MOGWO is used to obtain feature subsets from different datasets. Then machine learning models like KNN, random forest and logistic regression are used to obtain the accuracy results and comparison of the results is performed. The outputs from the 6 different datasets using MOGWO along with the machine learning models have been reviewed and summarized. The paper is concluded by mentioning the summary conclusion of MOGWO.
doi:10.5281/zenodo.4743482 fatcat:vqkuridbw5d3bla5t2wrqd2f2a