Reversing F-structure Rewriting for Generation from Meaning Representations
نویسندگان
چکیده
We describe the design of an LFG-based generation system that provides a framework for empirical studies on the choice among grammatical paraphrases (i.e. syntactic alternations), as an effect of interacting soft constraints. To be able to study the relevant variation, we extend the XLE generation architecture so it no longer departs from standard f-structures, but from a more abstract level of (meaning) representation. This representation is constructed by means of XFR term-rewrite rules. We discuss the design of the meaning representation in the light of the surface realisation task. In particular, we address the problem of obtaining a transfer grammar that reverses meaning construction, taking into account the generation performance.
منابع مشابه
Semantics-based Question Generation and Implementation
This paper presents a question generation system based on the approach of semantic rewriting. State-of-the-art deep linguistic parsing and generation tools are employed to map natural language sentences into their meaning representations in the form of Minimal Recursion Semantics (mrs) and vice versa. By carefully operating on the semantic structures, we obtain a principled way of generating qu...
متن کاملContextual Hyperedge Replacement Grammars for the Generation of Abstract Meaning Representations
We attempt to describe the properties of semantic data in the form of graphs by using a graph grammar. In particular, we construct a contextual hyperedge replacement grammar (CHRG) that can generate structurally correct abstract meaning representations (AMRs) over a limited domain of semantic concepts and relations. Moreover, the CHRG can generate minimal AMRs in the sense that none of the piec...
متن کاملAligning Formal Meaning Representations with Surface Strings for Wide-Coverage Text Generation
Statistical natural language generation from abstract meaning representations presupposes large corpora consisting of text–meaning pairs. Even though such corpora exist nowadays, or could be constructed using robust semantic parsing, the simple alignment between text and meaning representation is too coarse for developing robust (statistical) NLG systems. By reformatting semantic representation...
متن کاملA Class of Rewriting Rules and Reverse Transformation for Rule-based Equivalent Transformation
In the rule-based equivalent transformation (RBET) paradigm, where computation is based on meaning-preserving transformation of declarative descriptions, a set of rewriting rules is regarded as a program. The syntax for a large class of rewriting rules is determined. The incorporation of meta-variables of two different kinds enables precise control of rewriting-rule instantiations. As a result,...
متن کاملIssues in Generating Text from Interlingua Representations
Multi-lingual generation starts from non-linguistic content representations for generating texts in different languages that are equivalent in meaning. In contrast, cross-lingual generation is based on a language-neutral content representation which is the result of a linguistic analysis process. Non-linguistic representations do not reflect the structure of the text. Quite differently, languag...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2010