Media-driven Real-time Content Management Framework and its Application to In-Class Thinking Support System

In this paper, we present a media-driven real-time content management framework for interconnection between real world media and legacy media contents. Recently, it is easier to install various sensors in the real world, and a large amount of various kinds of media data are generated from these sensors. It is important to interconnect and utilize legacy media contents scattered on the Internet from media data generated in real time from the real world. The framework consists of four modules, “acquisition,” “extraction,” “selection,” and “retrieval.” By these four modules, it is possible to semantically interconnect real-world media data and legacy media contents. We apply our framework and realize thinking support system in classes. This system acquires teacher's voice in the class in real time, retrieves and presents legacy media contents corresponding to contents of the class on a screen. By this system, it is possible to show legacy media contents in real time on a screen according to the contents of the class without preparing in advance. This system is utilized by students to understand contents of the class and expand the ideas.