Application of binary grey wolf optimization algorithm to binary optimization problems

Metaheuristic algorithms are recommended and frequently used methods for solving optimization problems. Today, it has been adapted to many challenging problems and its successes have been identified. Grey Wolf Optimization Algorithm (GWO) is one of the next generation metaheuristic methods. Due to its advantages, GWO has been proposed for the solution of many different problems since the day it emerged and has proven its success. In this thesis, a new variant of GWO, Binary Dynamic Grey Wolf Optimization Algorithm (IDGKO), is proposed for the solution of binary optimization problems. The difference of IDGKO from other binary GWO methods is that it uses the XOR logic-gate-based operator to convert the continuous search space to binary search space and is based on the dynamic coefficient method developed to determine the effect of the three dominant individuals (alpha, beta and delta) in the original GWO on the solution quality. that is. IDGKO is intended to be a simple, feasible and successful method that strikes a balance between local search and global search in solving binary optimization problems. In order to determine the success and accuracy of the proposed IDGKO, tests were carried out on the 0-1 backpack problem (0-1 KP), which is a member of the feature selection (FS) and NP-Hard problems, which are frequently mentioned in the literature. FS is one of the basic preprocessing steps in data mining and is among the difficult binary optimization problems. FS is the process of determining the subset that can best represent the dataset by removing features that have little impact from a given dataset without affecting performance and accuracy. 0-1 KP has an active role in many different areas and its main purpose is to use resources efficiently with high profit and low cost. The performance of the proposed method has been compared with the performance of many algorithms, including recent binary GWO algorithms and different metaheuristic methods for the related problem. In the experiments, many data sets of different sizes and features were used to determine the efficiency and consistency of the algorithm. As a result of the comparisons and statistical tests, the proposed IDGKO has proven to be an effective and successful method in line with its purpose