Using News Articles for Real-time Cross-Lingual Event Detection and Filtering

The written medium through which we commonly learn about relevant news are news articles. Since there is an abundance of news articles that are written daily, the readers have a common problem of discovering the content of interest and still not be overwhelmed with the amount of it. In this paper we present a system called Event Registry which is able to group articles about an event across languages and extract from the articles core event information in a structured form. In this way, the amount of content that the reader has to check is significantly reduced while additionally providing the reader with a global coverage of each event. Since all event information is structured this also provides extensive and fine-grained options for information searching and filtering that are not available with current news aggregators.