Many kinds of knowledge-based system would be easier to develop and maintain if domain experts (as opposed to knowledge engineers) were in a position to define and edit the knowled...
Building NLG systems, in particular statistical ones, requires parallel data (paired inputs and outputs) which do not generally occur naturally. In this paper, we investigate the ...
We present three systems for surface natural language generation that are trainable from annotated corpora. The first two systems, called NLG1 and NLG2, require a corpus marked on...
1 This paper reports on the SPOKESMAN natural language generation system, which is a domain independent text generator designed to incrementally produce text for an underlying appl...