Applications had grown into complex applications, where such complexity negatively impacts usability by owing to provide single static user interface that should meet all users. We focus on adaptive user interfaces, which automatically customize user interface without direct user's invitation. In our previous work we have developed a novel adaptive interface called Boulevard, a panel container with a user interface automatically personalized to an individual user. We present our work-in-progress focused on enhancing Boulevard by utilization of affective computing and eye tracking, since considering user's current emotional state and point of gaze may play important role for performance of adaptive user interface.