Relationship between Degree of Awareness of Tastes for Users and Quality of Solutions to be Evaluated in Interactive Evolutionary Computation

Abstract

Interactive evolutionary computation (IEC) is one type of evolutionary computation which includes a human user as an evaluation system. We previously proposed an experimental way for investigating if iterations of evaluation of a solution candidate evolved by EC brings deeper awareness of tastes to users. However, we have not been able to reveal if evaluation of not any solution candidate but one evolved by EC is a must. Therefore, in the paper we first propose an experimental way for investigating if evaluation of a solution candidate evolved by EC is a must for deeper awareness of tastes, and then, conduct an experiment in the proposed way to reveal that. The experimental results show that some participants obtain deeper awareness of their tastes from randomly generated solution candidates rather than ones evolved by EC and also that there is no statistically significant difference between the effect of evaluation of a solution candidate evolved by EC and that of a randomly generated solution candidate on deepening awareness of tastes.