Newsinessence: Summarizing online news topics

NewsInEssence (NIE) is a new delivery and summarization system under development at the University of Michigan, which gathers and recaps news items based on specifications and interests to online news readers. NIE searches across dozens of news sites to collect a group, or cluster, of related stories and then generates a summary of the entire cluster, highlighting its most important content. NIE provides NewsTroll component for the user to create news clusters for them, which allows the user to specify parameters, such as which sources to use, that can customize the retrieved cluster. NIE uses MEAD, a public-domain summarizer, to automatically precompute summaries at set compression rates. With NIE's tracking option, a user can request an update on a specific cluster to be sent directly to his or her box at a specific time. It enables the users to specify their choice of virtual anchors to deliver, in real time, news collected and summarized according to their preferences.