Extracting Information from Conference Announcements: High Recall, High Precision

Recall, High Precision Kevin Cheong Language Technology Group, Microsoft Research Institute School of MPCE, Macquarie University Sydney NSW 2109, Australia [email protected] Abstract Conference announcements are distributed widely each day via electronic mail to the research and industrial community. These conferences inform researchers, academics and the industry about the research and development (R & D) work performed in a particular eld of interest. There is a wealth of information contained in this multitude of conference announcements. The aim of this research is to extract essential and relevant information from conference announcements and to explore the technologies involved. In this paper we describe an architecture for a system we have developed that extracts relevant and useful information from conference announcement electronic mail messages, with a focus on achieving a high recall and precision rate. We also discuss the extent to which the success of this information extraction task depends on domain and world knowledge.