| Readers interested in the context of an event covered in the news, such as the announcement of a verdict in a legal trial, can benefit from easily finding out about the overall news situation, the trial, of which the event is a part. Guided by abstract models of news situation types including legal trials, corporate acquisitions, and kidnappings, Brussell supports readers by presenting a storyline for a situation and facts about its participants. It gathers this information by reading news articles about the situation and, in contrast to previous work in event-extraction, topic tracking and news summarization, Brussell is the first research system to extract and aggregate information from multiple news articles describing multiple component events of specific ongoing situations. We find that gathering situation information in this way significantly improves F1-measure performance in extracting the dates of events as more articles are read. |
