ESSENCE: a Protable Methodology for Acquiring Information Extraction Patterns

摘要

One important issue when constructing Information Extraction systems is how to obtian the knowledge needed for identifying relevant informaiton in a document. In most apporahces to this issue, the human expert interention is necessary in many steps of the acquisition process. In this paper we describe ESSENCE, a new methodolgoy that reduces significantly the need for human intervention. It is based on ELA, a new algorithm for acquiring informaiton extraction patterns.