Information Extraction (IE) has already reached the level of success at which Information Retrieval and Machine Translation (on differing measures, of course) have proved commercially viable. By general agreement, the main barrier to wider use and commercialization of IE is the relative inflexibility of the template concept: classic IE relies on the user having an already developed set of templates, as was the case with US Defence agencies from where the technology was largely developed (see below), and this is not generally the case. The intellectual and practical issue now is how to develop templates, their subparts (like named entities or NEs), the rules for filling them, and associated knowledge structures, as rapidly as possible for new domains and genres.
This paper discusses the quasi-automatic development and detection of templates, template-fillers, lexicons and knowledge structures for new IE domains and genres, using a combination of machine learning, linguistic resource extrapolation and human machine interface elicitation and feedback techniques.