Research on emotion has been done mainly in psychology. But recently, emotion has been dealt with in Kansei engineering and Kansei Information Processing and some human emotional models have been applied to industrial products. When we think about the interface between persons and robots, It is very important to use emotional expressions for the mutual communication. In this paper, we propose an emotional processing model using genetic algorithm for networks that change the emotional information from the database information to color information. This model has two function parts. First, we define the fundamental emotion groups that have grades. Next, we construct the emotion database and change the emotional information to HVC color information.