Shared task system description: frustratingly hard compositionality prediction

Authors:
Anders Johannsen;Hector Martinez Alonso;Christian Rishøj;Anders Søgaard
Affiliations:
University of Copenhagen;University of Copenhagen;University of Copenhagen;University of Copenhagen
Venue:
DiSCo '11 Proceedings of the Workshop on Distributional Semantics and Compositionality
Year:
2011

Citing 2
Cited 2

The S-Space package: an open source package for word space models

ACLDemos '10 Proceedings of the ACL 2010 System Demonstrations
Predicting the semantic compositionality of prefix verbs

EMNLP '10 Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing

Domain and function: a dual-space model of semantic relations and compositions

Journal of Artificial Intelligence Research
Unsupervised feature adaptation for cross-domain NLP with an application to compositionality grading

CICLing'13 Proceedings of the 14th international conference on Computational Linguistics and Intelligent Text Processing - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

We considered a wide range of features for the DiSCo 2011 shared task about compositionality prediction for word pairs, including COALS-based endocentricity scores, compositionality scores based on distributional clusters, statistics about wordnet-induced paraphrases, hyphenation, and the likelihood of long translation equivalents in other languages. Many of the features we considered correlated significantly with human compositionality scores, but in support vector regression experiments we obtained the best results using only COALS-based endocentricity scores. Our system was nevertheless the best performing system in the shared task, and average error reductions over a simple baseline in cross-validation were 13.7% for English and 50.1% for German.