From Cases to Rules and Vice Versa: Robust Practical Parsing With Analogy
Alex Chengyu Fang · International Workshop/Conference on Parsing Technologies · 2000
This article describes the architecture of the Survey Parser and discusses two major components related to the analogy-based parsing of unrestricted English. Firstly, it discusses the automatic generation of a large declarative formal grammar from a corpus that has been syntactically analysed. Secondly, it describes analogy-based parsing that employs both the automatically learned rules and the database of cases to determine the syntactic structure of the input string. Statistics are presented to characterise the performance of the parsing system.