Graph parsing with s-graph grammars

Groschwitz, Jonas; Koller, Alexander; Teichmann, Christoph

doi:10.3115/v1/p15-1143

Cited by 27 publications

(27 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…the rules of the wRTG are only calculated by need; see e.g. (Groschwitz et al, 2015;Groschwitz et al, 2016). Obviously, Alto cannot be as efficient for well-established tasks like PCFG parsing as a parser that was implemented and optimized for this specific grammar formalism.…”

Section: Algorithms In Altomentioning

confidence: 99%

“…Nonetheless, Alto is fast enough for practical use with treebank-scale gramars, and for less mainstream grammar formalisms can be faster than specialized implementations for these formalisms. For instance, Alto is the fastest published parser for Hyperedge Replacement Grammars (Groschwitz et al, 2015). Alto contains multiple algorithms for computing the intersection and inverse homomorphism of RTGs, and a user can choose the combination that works best for their particular grammar formalism (Groschwitz et al, 2016).…”

Section: Algorithms In Altomentioning

confidence: 99%

See 1 more Smart Citation

Proceedings of the Software Demonstrations of the 15th Conference of the European Chapter of the Association for Computational Linguistics

2017

View full text Add to dashboard Cite

ii Introduction This volume contains the contributions accepted for Software Demonstrations at the EACL 2017 conference which takes place from 4th to 7th of April 2017 in Valencia, Spain.The Demonstrations program primary aim is to present the software systems and solutions related to all areas of computational linguistics and natural language processing. This program encourages the early exhibition of research prototypes, but also includes interesting mature systems.The number of submissions for software demonstrations exceeded, again, the number of available slots. This confronted us with the situation where we were forced to introduce a strict selection process. The selection process was completed following the predefined criteria of relevance for EACL 2017 conference: novelty, overall usability, and applicability to more than one linguistic level and to more than one computational linguistic field/task. We would like to thank the general chairs and the local organizers, without whom it would have been impossible to put together such a strong demonstrations program.We hope you will find this proceedings useful and that the software systems presented in this proceedings will inspire new ideas and approaches in your future work. AbstractIn semantic parsing, natural language questions map to meaning representation language (MRL) expressions over some fixed vocabulary of predicates. To do this reliably, one must guarantee that for a wide class of natural language questions (the so called semantically tractable questions), correct interpretations are always in the mapped set of possibilities. Here we demonstrate the system COVER which significantly clarifies, revises and extends the notion of semantic tractability. COVER is written in Python and uses NLTK.

show abstract

Section: Algorithms In Altomentioning

confidence: 99%

Section: Algorithms In Altomentioning

confidence: 99%

Proceedings of the Software Demonstrations of the 15th Conference of the European Chapter of the Association for Computational Linguistics

2017

View full text Add to dashboard Cite

show abstract

“…AMR parsing systems that focus on modeling the graph aspect of the AMR includes JAMR (Flanigan et al, 2014(Flanigan et al, , 2016aZhou et al, 2016), which treats AMR parsing as a procedure for searching for the Maximum Spanning Connected Subgraphs (MSCGs) from an edge-labeled, directed graph of all possible relations. Parsers based on Hyperedge Replacement Grammars (HRG) (Chiang et al, 2013;Björklund et al, 2016;Groschwitz et al, 2015) put more emphasis on modeling the formal properties of the AMR graph. One practical implementation of HRG-based parsing is that of (Peng et al, 2015;Peng and Gildea, 2016).…”

Section: Related Workmentioning

confidence: 99%

Getting the Most out of AMR Parsing

Wang

Xue

2017

Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

This paper proposes to tackle the AMR parsing bottleneck by improving two components of an AMR parser: concept identification and alignment. We first build a Bidirectional LSTM based concept identifier that is able to incorporate richer contextual information to learn sparse AMR concept labels. We then extend an HMM-based word-to-concept alignment model with graph distance distortion and a rescoring method during decoding to incorporate the structural information in the AMR graph. We show integrating the two components into an existing AMR parser results in consistently better performance over the state of the art on various datasets.

show abstract

“…A is uniquely determined by its boundary representation (Lautemann, 1990;Chiang et al, 2013;Groschwitz et al, 2015), which consists of (a) the pair (ports H , ϕ(ports H )), (b) the set of boundary edges E B H of H consisting of all e ∈ E H such that att H (e) ∩ [ports H ] = ∅, and (c) a function att :…”

mentioning

confidence: 99%

EM-Training for Weighted Aligned Hypergraph Bimorphisms

Drewes¹,

Gebhardt²,

Vogler³

2016

Proceedings of the SIGFSM Workshop on Statistical NLP and Weighted Automata

View full text Add to dashboard Cite

We develop the concept of weighted aligned hypergraph bimorphism where the weights may, in particular, represent probabilities. Such a bimorphism consists of an R ≥0 -weighted regular tree grammar, two hypergraph algebras that interpret the generated trees, and a family of alignments between the two interpretations. Semantically, this yields a set of bihypergraphs each consisting of two hypergraphs and an explicit alignment between them; e.g., discontinuous phrase structures and nonprojective dependency structures are bihypergraphs. We present an EM-training algorithm which takes a corpus of bihypergraphs and an aligned hypergraph bimorphism as input and generates a sequence of weight assignments which converges to a local maximum or saddle point of the likelihood function of the corpus.

show abstract

Graph parsing with s-graph grammars

Cited by 27 publications

References 11 publications

Proceedings of the Software Demonstrations of the 15th Conference of the European Chapter of the Association for Computational Linguistics

Proceedings of the Software Demonstrations of the 15th Conference of the European Chapter of the Association for Computational Linguistics

Getting the Most out of AMR Parsing

EM-Training for Weighted Aligned Hypergraph Bimorphisms

Contact Info

Product

Resources

About