The use of statistical techniques has revolutionized the field of computational linguistics over the last ten years. In this talk we review the state of the art in programs that can assign semantic annotations to general text. These programs typically start by producing a parse tree for the sentences of the text. There are now several programs that can produce reasonably accurate complete parses for naturally occurring English (e.g., all of the sentences on the front page of today's New York Times).
展开▼