An Empirical Evaluation of AMR Parsing for Legal Documents

摘要

Many approaches have been proposed to tackle the problem of Abstract Meaning Representation (AMR) parsing, help solving various natural language processing issues recently. In our paper, we provide an overview of different methods in AMR parsing and their performances when analyzing legal documents. We conduct experiments of different AMR parsers on our annotated dataset extracted from the English version of Japanese Civil Code. Our results show the limitations as well as open a room for improvements of current parsing techniques when applying in this non-trivial domain.