Multilingual semantic role labeling

Abstract

This paper describes our contribution to the semantic role labeling task (SRL-only) of the CoNLL-2009 shared task in the closed challenge (Hajič et al., 2009). Our system consists of a pipeline of independent, local classifiers that identify the predicate sense, the arguments of the predicates, and the argument labels. Using these local models, we carried out a beam search to generate a pool of candidates. We then reranked the candidates using a joint learning approach that combines the local models and proposition features. To address the multilingual nature of the data, we implemented a feature selection procedure that systematically explored the feature space, yielding significant gains over a standard set of features. Our system achieved the second best semantic score overall with an average labeled semantic F1 of 80.31. It obtained the best F1 score on the Chinese and German data and the second best one on English.

Low-Resource Semantic Role Labeling

“…Previous related approaches to semantic role labeling include joint classification of semantic arguments (Toutanova et al, 2005;Johansson and Nugues, 2008), latent syntax induction (Boxwell et al, 2011;Naradowsky et al, 2012), and feature engineering for SRL (Zhao et al, 2009;Björkelund et al, 2009). Toutanova et al (2005) introduced one of the first joint approaches for SRL and demonstrated that a model that scores the full predicateargument structure of a parse tree could lead to significant error reduction over independent classifiers for each predicate-argument relation.…”

“…Models for SRL have increasingly come to rely on an array of NLP tools (e.g., parsers, lemmatizers) in order to obtain state-of-the-art results (Björkelund et al, 2009;Zhao et al, 2009). Each tool is typically trained on hand-annotated data, thus placing SRL at the end of a very highresource NLP pipeline.…”

Improving Chinese Semantic Role Labeling using High-quality Surface and Deep Case Frames

“…We basically use the feature set proposed by Björkelund et al (2009). During the prediction, there are some predicates which have not been seen in the training data.…”

“…This type of framework has the ability to use SRL information in syntactic parsing for improvement, but needs a much larger search space for decoding. The other type is called SRLonly task (Zhao et al, 2009;Björkelund et al, 2009), which uses automatic morphological and syntactic information as the input in order to judge which token plays what kind of semantic role. Our work focuses on the second category of SRL.…”

“…Our work focuses on the second category of SRL. Our framework is based on those used by Björkelund et al (2009) and Yang and Zong (2014).…”