An empirical evaluation of AMR parsing for legal documents
This work addresses the challenge of applying AMR parsing to legal texts, which is an incremental contribution to domain-specific NLP.
The paper evaluated existing Abstract Meaning Representation (AMR) parsing methods on legal documents using a dataset from the Japanese Civil Code, revealing limitations and opportunities for improvement in this complex domain.
Many approaches have been proposed to tackle the problem of Abstract Meaning Representation (AMR) parsing, helps 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 complicated domain.