Context-Adaptive User Interface in Ubiquitous Home Generated by Bayesian and Action Selection Networks

Recent home theater system requires for users to control various devices such as TV, audio equipment, DVD and video players, and set-top box simultaneously. To obtain the services that a user wants in this situation, user should know the functions and positions of the buttons in several remote controllers. Since it is usually difficult to manipulate them and the number of the devices that we can control increases, we get to confuse more as the ubiquitous home environment is realized. Moreover, there are a lot of mobile and stationary controller devices in ubiquitous computing environment, so that user interface should be adaptive in selecting the functions that user wants and in adjusting the features of UI to fit in a specific controller. To implement the user and controller adaptive interface, we model the ubiquitous home environment and use the modeled context and device information. We have used Bayesian network to get the degree of necessity in each situation. Action selection network uses predicted user situation and necessary devices, and it selects necessary functions in current situation. Selected functions are used to construct adaptive interface for each controller using presentation template. For experiments, we have implemented ubiquitous home environment and generated controller usage log in this environment. We have confirmed the BN predicted user requirements effectively as evaluating the inferred results of controller necessity based on generated scenario. Finally, comparing the adaptive home UI with the fixed one by 14 subjects, we confirm that the generated adaptive UI is more useful for general tasks than the fixed UI.