On-line knowledge embedding for an interactive EC-based montage system

We propose an on-line knowledge embedding method for interactive evolutionary computation (EC) to reduce human fatigue. On-line knowledge embedding is a method that a user explicitly uses searching hints that the user obtained during an interactive EC operation for the next search. To evaluate this method, we developed two interactive EC-based montage systems. The first system embedded the user's intention in an EC search and the second did not. The result of subjective tests and other evaluation measurements have shown that the proposed method realized a faster convergence than ordinary interactive EC searches.