Transformation-based information extraction using learned meta-rules

Abstract

Information extraction (IE) is a form of shallow text understanding that locates specific pieces of data in natural language documents. Although automated IE systems began to be developed using machine learning techniques recently, the performances of those IE systems still need to be improved. This paper describes an information extraction system based on transformation-based learning, which uses learned meta-rules on patterns for slots. We plan to empirically show these techniques improve the performance of the underlying information extraction system by running experiments on a corpus of IT resumé documents collected from Internet newsgroups. © Springer-Verlag Berlin Heidelberg 2005.