Building a Generation Knowledge Source using Internet-Accessible Newswire

Building a Generation Knowledge Source using Internet-Accessible Newswire Dragomir R. Radev and Kathleen R. McKeown Department of Computer Science Columbia University New York, NY 10027 {radev,kathy}@cs.columbia.edu Abstract In this paper, we describe a method for automatic creation of a knowledge source for text generation using information ex- traction over the Internet. We present a prototype system called PROFILE which uses a client-server architecture to ex- tract noun-phrase descriptions of enti- ties such as people, places, and organiza- tions. The system serves two purposes: as an information extraction tool, it al- lows users to search for textual descrip- tions of entities; as a utility to generate functional descriptions (FD), it is used in a functional-unification based generation system. We present an evaluation of the approach and its applications to natural language generation and summarization. 1 Introduction In our work to date on news summarization at Columbia University (McKeown and Radev, 1995; Radev, 1996), information is extracted from a se- ries of input news articles (MUC, 1992; Grishman et al., 1992) and is analyzed by a generation com- ponent to produce a summary that shows how per- ception of the event has changed over time. In this summarization paradigm, problems arise when in- formation needed for the summary is either miss- ing from the input article(s) or not extracted by the information extraction system. In such cases, the information may be readily available in other current news stories, in past news, or in online databases. If the summarization system can find the needed information in other online sources, then it can produce an improved summary by merging information from multiple sources with information extracted from the input articles. In the news domain, a summary needs to refer to people, places, and organizations and provide descriptions that clearly identify the entity for the reader. Such descriptions may not be present in the original text that is being summarized. For ex- ample, the American pilot Scott O'Grady, downed in Bosnia in June of 1995, was unheard of by the American public prior to the incident. If a reader tuned into news on this event days later, descrip- tions from the initial articles may be more useful. A summarizer that has access to different descrip- tions will be able to select the description that best suits both the reader and the series of articles be- ing summarized. In this paper, we describe a system called PROFILE that tracks prior references to a given entity by extracting descriptions for later use in summarization. In contrast with previous work on information extraction, our work has the following features: • It builds a database of profiles for entities by storing descriptions from a collected corpus of • past news. • It operates in real time, allowing for connec- tions with the latest breaking, online news to extract information about the most recently mentioned individuals and organizations. • It collects and merges information from dis- tributed sources thus allowing for a more com- plete record of information. • As …

Content truncated to 500 words