Discriminative Reranking of Discourse Parses Using Tree Kernels
نویسندگان
چکیده
In this paper, we present a discriminative approach for reranking discourse trees generated by an existing probabilistic discourse parser. The reranker relies on tree kernels (TKs) to capture the global dependencies between discourse units in a tree. In particular, we design new computational structures of discourse trees, which combined with standard TKs, originate novel discourse TKs. The empirical evaluation shows that our reranker can improve the state-of-the-art sentence-level parsing accuracy from 79.77% to 82.15%, a relative error reduction of 11.8%, which in turn pushes the state-of-the-art documentlevel accuracy from 55.8% to 57.3%.
منابع مشابه
Discriminative Reranking for Natural Language Parsing
This paper considers approaches which rerank the output of an existing probabilistic parser. The base parser produces a set of candidate parses for each input sentence, with associated probabilities that define an initial ranking of these parses. A second model then attempts to improve upon this initial ranking, using additional features of the tree as evidence. We describe and compare two appr...
متن کاملA Reranking Model for Discourse Segmentation using Subtree Features
This paper presents a discriminative reranking model for the discourse segmentation task, the first step in a discourse parsing system. Our model exploits subtree features to rerank Nbest outputs of a base segmenter, which uses syntactic and lexical features in a CRF framework. Experimental results on the RST Discourse Treebank corpus show that our model outperforms existing discourse segmenter...
متن کاملCoarse-to-Fine n-Best Parsing and MaxEnt Discriminative Reranking
Discriminative reranking is one method for constructing high-performance statistical parsers (Collins, 2000). A discriminative reranker requires a source of candidate parses for each sentence. This paper describes a simple yet novel method for constructing sets of 50-best parses based on a coarse-to-fine generative parser (Charniak, 2000). This method generates 50-best lists that are of substan...
متن کاملForest Reranking: Discriminative Parsing with Non-Local Features
Conventional n-best reranking techniques often suffer from the limited scope of the n-best list, which rules out many potentially good alternatives. We instead propose forest reranking, a method that reranks the packed forest of exponentially many parses. Although exact inference is intractable with non-local features, we present an approximation algorithm that makes discriminative training pra...
متن کاملFeatures for Phrase-Structure Reranking from Dependency Parses
Radically different approaches have been proved to be effective for phrase-structure and dependency parsers in the last decade. Here, we aim to exploit the divergence in these approaches and show the utility of features extracted from the automatic dependency parses of sentences for a discriminative phrase-structure parser. Our experiments show a significant improvement over the state-of-the-ar...
متن کامل