Supporting for Real-World Tasks: Producing Summaries of Scientific Articles Tailored to the Citation Context

Abstract

The amount of scientific material available electronically is forever increasing. This makes reading the published literature, whether to stay up-to-date on a topic or to get up to speed on a new topic, a difficult task. Yet, this is an activity in which all researchers must be engaged on a regular basis. Based on a user requirements analysis, we developed a new research tool, called the Citation-Sensitive In-Browser Summariser (CSIBS), which supports researchers in this browsing task. CSIBS enables readers to obtain information about a citation at the point at which they encounter it. This information is aimed at enabling the reader to determine whether or not to invest the time in exploring the cited article further, thus alleviating information overload. CSIBS builds a summary of the cited document, bringing together meta-data about the document and a citation-sensitive preview that exploits the citation context to retrieve the sentences from the cited document that are relevant at this point. In this talk, I will briefly present our user requirements analysis, then describe the system and, finally, discuss the observations from an initial pilot study. We found that CSIBS facilitates the relevancy judgment task, by increasing the users’ self-reported confidence in making such judgements.