Publication Date



Open access

Embargo Period


Degree Type


Degree Name

Doctor of Philosophy (PHD)


Electrical and Computer Engineering (Engineering)

Date of Defense


First Committee Member

Miroslav Kubat

Second Committee Member

Kamal Premaratne

Third Committee Member

Michael Scordilis

Fourth Committee Member

Nigel M. John

Fifth Committee Member

Sundararaman G. Gopalakrishnan


The Genetic Algorithm (GA) is a popular approach to search and optimization that has been applied to hundreds of real-world optimization problems across numerous domains of science. The GA describes an iterative search process that seeks to improve the quality of an initially random set of solutions with respect to some user-defined optimization criteria. The components of this iterative search process mimic Darwinian biological evolutionary processes such as mating, recombination, mutation, and survival of the fittest. Over the years, researchers have attempted to improve various components of the GA search process. However, the impact of the mating strategy, which determines how existing solutions to a problem are paired during the genetic search process to generate new and better solutions, has so far been neglected in the rich and vast GA literature. In this work, five novel mating strategies inspired from the Darwinian evolutionary principle of "opposites-attract" are proposed to speed up the GA search process. The impact of the proposed mating strategies on the GA’s performance is tested on two well-established and complex testbed optimization problems from the domain of supervised classification: 1) the 1-NN Tuning problems, and 2) the Optimal Decision Forests problem. The results from rigorous experiments with various UCI data sets reveal that the proposed mating strategies both accelerate the GA search and lead to the discovery of better solutions. Moreover, these improvements come at the cost of only negligible additional computational overhead.


Genetic Algorithms; Mating Strategies; Machine Learning; Optimization; Data Mining; Supervised Classification