UFJF - Universidade Federal de Juiz de Fora

Computational Semantics with Frames and Constructions

You are at: Home > Theme sessions > Computational Semantics with Frames and Constructions


Miriam R. L. Petruck | International Computer Science Institute

Nathan Schneider | Georgetown University


The goal of this theme session is to bring together (frame) semanticists, (construction) grammarians, and computational linguists interested in advancing the role(s) of frames and constructions in computational semantics—or conversely, advancing the role of computational approaches in the characterization of frames and constructions. We seek contributions that address the following question broadly construed: How does the work inform the understanding of computing the meaning of a frame  or construction ?

We seek computational work involving semantic frames  (Fillmore 2012, inter alia ) and/or grammatical constructions , i.e. form-meaning pairings (Fillmore et al. 2012). We are especially interested in novel approaches to familiar natural language processing techniques and tasks. Submissions should use theoretical or empirical methods to address questions such as: How do frames/constructions figure into computational models of human processing, production, and acquisition of language? How can natural language processing systems exploit frames and constructions for semantic interpretation and other tasks? Topics of interest include, but are not limited to:

●  semantic role labeling

●  semantic parsing with frames and/or constructions

●  models of language acquisition & change

●  lexicon/grammar induction with frames and/or constructions

●  selectional preferences

●  disambiguation of semantic relations (e.g. in compounds or other multiwords)

●  events, temporal structure, causality

●  script/narrative induction & reasoning

●  multiword expressions & idioms

●  construction detection/disambiguation/parsing

●  computational construction grammar formalisms (e.g. Embodied Construction Grammar, Fluid Construction Grammar, Sign-Based Construction Grammar, HPSG)

●  grounded semantics (e.g. vision, space & time, actions/robotics, neuroscience)

●  frames or constructions in semantic similarity, paraphrasing, entailment, generation, discourse processing, translation, sentiment analysis, and other applications

This theme session is specifically designed for work that involves algorithms, formalisms, or systems, not (exclusively) data resources or annotation schemes.

Keywords: Computational Semantics, Frame Semantics, Construction Grammar, frames, constructions

Accepted Papers: 


Oral Presentations:


OmniGraph: Semantic Frame Representation and Graph Kernel Learning

B. Xie and R. J. Passonneau

Vector Space for Semantic Roles: Distances and Semantic Maps

O. Lyashevskaya and E. Kashkin

Unsupervised Induction of Shallow Semantic Representations with Feature-Rich Models

D. Marcheggiani, E. Khoddam, and I.Titov

Computational Corpus-based Metaphor Analysis with Frames and Constructions

E. Stickles and E. K. Dodge

Proto-properties of Definiteness and Grammatical Constructions

A. Bhatia and C.-C. Lin 

CARMA: Constructional Analysis using Relations among Multiple Attribute-Value Matrices

T. Torrent, E. Matos, A. Silva, L. Meireles, A. Diniz, V. Almeida and T. Tavares

Improving Frame Semantics Parsing via Dependency Path Embeddings

M. Roth


Poster Presentations:


Embodied Construction Grammar Meaning Representation of Frames and Constructions: Caused_motion.

E. K. Dodge and M. R. L. Petruck

Larger than Words – Construction Disambiguation with Swedish FrameNet and Constructicon

A. Ehrlemark and R. Johansson

Semantic Relation Disambiguation in Swedish Compounds with Swedish FrameNet and SALDO

K. Hedberg and R. Johansson

Frame-based resolution of coreference in the case of the highly inflected language

M. Ogrodniczuk, M. Zawisławska and F. M. Tyers

Distributional Embeddings and Diachronic Construction Grammar

S. Budts, M. Kestemont, and P. Petré

Grammatical Framework for Implementing Multilingual Frames and Constructions

N. Gruzitis, D. Dannélls, and A., Ranta