Creation of Delicious Mixed Juices for Multiple Users based on Distributed Interactive Genetic Algorithm

Creating media content suited to many users' feelings is essential in a situation with multiple users' participation. Interactive Evolutionary Computation (IEC) is a search method that finds good solutions; however, it is originally for each user. Previous studies proposed that IEC creates mixed juices for two users. Allowing many users' participation contributes to finding good solutions suited to many users and will support developing good products. This study aims to propose an IEC in which trio of users can participate. The elite Genetic Algorithm individuals were exchanged between the users in this method. A concrete system based on the proposed method was constructed, and the experiment was conducted with the system. Fifteen students participated in the experiment. Changes in the fitness values and search space were investigated. Furthermore, the exchanged individuals were compared with the individuals created by the examinee's population by their fitness values. The statistical analyses showed that the proposed method could find good mixed juices for the trio of examinees, and the exchange of the elite individuals had the potential for an efficient search.