Extracting Information from Text

Abstract

A domain independent information extraction system can be built if it can include a self modifying feature that enables it to automatically adapt to new domains. The theory is that the user should know the peculiarities involved and be able to quickly train the system to gather the specific target facts. This is done by scanning a series of example articles using a special text processing interface and allowing the user to lead the system through the desired retrieval steps. Then the system acquires the generalized patterns and applies them to any large database of text to extract the information of interest.