Survey on Event tracking and Event Evolution

Due to the popularity of the Internet, most news stories have electronic versions published on newswires. Retrieving news of the same topic from multiple sources and keeping information updated becomes more convenient and easier. Techniques that are capable of extracting the underlying structure of the news events are desired. They are helpful to understand the evolution of events on the same topic. This paper surveys the different information retrieval techniques such as Topic detection and Tracking (TDT), event tracking and event evolution approaches. TDT is an active research area in recent years. TDT have different techniques on detecting news topics and tracking news stories for a news topic. TDT techniques have been attempting to detecting or clustering news stories into event, without defining or interpreting the association between these events. There also event evolution is a new concept developed recently, that was the first to conduct investigation on event. Event evolution is the transition development process of related events within the same topic. An event evolution graph is constructed to present events relationship.