Bayesian TV: Emergency Information Media by Automating Coverage, Distribution and Editing
Finished by March 31, 2013
Research Organization of Information and Systems, National Institute of Informatics, Digital Content and Media Sciences Research Division, Associate Professor
This project aims at making "Bayesian TV" for emergency information where real-time information is relevant. The TV is based on a temporal Bayesian model and automates coverage, distribution and editing. The present television system has a drawback on personalizing programs, while the present Web has a drawback on requiring an active information seeking process. Hence we design a "push" media that automatically provides relevant information using a television-like interface, and our goal is to start "Digital Typhoon TV" on the Web providing typhoon information to the general public.