Effects of unnatural selection on genetic algorithms

The genetic algorithm is one of the heuristic algorithms that have been used for many years in problems where optimization techniques are needed in the solution. This algorithm, which is based on simulating evolutionary processes, has been used in different ideal solution problems and has contributed to the literature with the solutions they have produced. Genetic algorithms based on the selection and evolutionary process in nature aim to reach ideal solution sets by producing different population groups. Gene transfer and genome editing in genetics are among the fields of study that have increased their popularity in recent years. The development of new techniques for gene transfer between living things has paved the way for alternatives other than classical natural selection. In this study, it is aimed to adapt the transfer of the genetic structure of living things from another living thing to genetic algorithms and to see what kind of changes this process will cause in the operation of the algorithm. The working principle of genetic algorithm applications is to produce two new individuals as a result of combining the gene fragments obtained by the crossover method from two individuals. The study is based on the continuation of the classical genetic algorithm processes of two new individuals, which will be created as a result of the inclusion of a third individual in the crossover process. In order to compare the new crossover method produced with the classical method, the Knapsack Problem, which is widely used in the literature, was preferred. The classical method and the new method have been compared in terms of success rates and solution generation rates, and the comparison results have been presented in tabular form. When the results obtained in the study were examined, it was seen that the proposed new crossover method had a higher success rate compared to the classical method and reached the result in earlier iterations.