Dynamic gain estimation in ambient media services

Abstract

This paper presents a mechanism for dynamically estimating the gain in ambient media services. Gain estimation is an attempt to measure the extent to which a certain media service is useful for a user in a particular context. Such a measure may provide invaluable support for personalizing a user's ambient environment through the selection of timely, context-aware and interesting media services. The proposed method considers multiple factors including the user's profile, context, media reputation, and the interaction history to dynamically estimate the gain of a service. The gain estimation method has been incorporated in a prototype smart mirror system in a smart environment setting. Experimental results demonstrate the suitability of the proposed method.