Frame-Semantic Parsing

Abstract

Frame semantics is a linguistic theory that has been instantiated for English in the FrameNet lexicon. We solve the problem of frame-semantic parsing using a two-stage statistical model that takes lexical targets (i.e., content words and phrases) in their sentential contexts and predicts frame-semantic structures. Given a target in context, the first stage disambiguates it to a semantic frame. This model uses latent variables and semi-supervised learning to improve frame disambiguation for targets unseen at training time. The second stage finds the target's locally expressed semantic arguments. At inference time, a fast exact dual decomposition algorithm collectively predicts all the arguments of a frame at once in order to respect declaratively stated linguistic constraints, resulting in qualitatively better structures than naïve local predictors. Both components are feature-based and discriminatively trained on a small set of annotated frame-semantic parses. On the SemEval 2007 benchmark data set, the approach, along with a heuristic identifier of frame-evoking targets, outperforms the prior state of the art by significant margins. Additionally, we present experiments on the much larger FrameNet 1.5 data set. We have released our frame-semantic parser as open-source software.

Controlled Natural Language

“…For this use case, the information extraction process, i.e., the semantic parsing of input texts cannot be approached by CNL: large-scale media monitoring is not limited to a particular domain, and the input sources vary from newswire texts to radio and TV transcripts to usergenerated content in social networks. Robust machine learning techniques are necessary instead to map the arbitrary input sentences to their meaning representation in terms of PropBank and FrameNet [7], or the emerging Abstract Meaning Representation, AMR [8], which is based on PropBank with named entity recognition and linking via DBpedia [9]. AMR parsing has reached 67% accuracy (the F 1 score) on opendomain texts, which is a level acceptable for automatic summarization [10].…”

Chapter 1. Introduction

“…Section 6.3.1 is primarily devoted to describing the utility of the relations for humans using FrameNet as a reference, Section 6.3.2 is of use to both humans and automatic programs, whereas the other sections are of more interest to developers intending to use FrameNet for computational purposes. Virtually any computational use of the FrameNet relation information relies on and presupposes semantic parsing of texts-a process not discussed here, but well covered in several publications (e.g., Das et al (2014) and Palmer et al (2010)). …”

Towards knowledge modeling and manipulation technologies: A survey

“…Frames represent generalizations over groups of words which illustrate equivalent situations, similar set of roles and related syntactic behaviour (Martínez-Santiago et al, 2015;O'Hara & Wiebe, 2009). In the theory of frame semantics, the roles or common situations which describes a frame are called frame elements (Das et al, 2014;Pimentel et al, 2012). The association between a word form and its meaning is referred to as a lexical unit (Zhang et al, 2015).…”

“…FN corpus can be assessed using the Natural Language Toolkit (NLTK) in Python (Garrette & Klein, 2009). The data in FN has been used to develop automatic semantic role labellers (Croce & Basili, 2011;Erk & Pado, 2006;Giannone, 2013;Gildea & Jurafsky, 2002;Padó & Lapata, 2009) and frame-semantic parsers (FSP) (Das et al, 2014). Other end-user applications for FN includes Question answering (QA) (Ofoghi et al, 2008a(Ofoghi et al, , 2008bSinha, 2008) and information extraction (IE) (Mohit & Narayanan, 2003;Scaiano & Inkpen, 2009).…”