A Linguistically-motivated 2-stage Tree to Graph Transformation

Abstract : We propose a new model for transforming dependency trees into target graphs, relying on two distinct stages. During the first stage, standard local tree transformation rules based on patterns are applied to collect a first set of constrained edges to be added to the target graph. In the second stage, motivated by linguistic considerations, the constraints on edges may be used to displace them or their neighbour edges upwards, or to build new mirror edges. The main advantages of this model is to simplify the design of a transformation scheme, with a smaller set of simpler local rules for the first stage, and good properties of termination and confluence for the second level.
Document type :
Conference papers
Complete list of metadatas

Cited literature [16 references]  Display  Hide  Download

https://hal.inria.fr/hal-00765422
Contributor : Eric Villemonte de la Clergerie <>
Submitted on : Friday, December 14, 2012 - 3:48:09 PM
Last modification on : Friday, May 3, 2019 - 1:41:52 AM
Long-term archiving on : Sunday, December 18, 2016 - 2:23:55 AM

File

rewriting.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00765422, version 1

Citation

Corentin Ribeyre, Djamé Seddah, Éric Villemonte de la Clergerie. A Linguistically-motivated 2-stage Tree to Graph Transformation. TAG+11 - The 11th International Workshop on Tree Adjoining Grammars and Related Formalisms - 2012, INRIA, Sep 2012, Paris, France. ⟨hal-00765422⟩

Share

Metrics

Record views

383

Files downloads

180