Prepositional Phrase Attachment over Word Embedding Products

Pranava Swaroop Madhyastha, Xavier Carreras, Ariadna Quattoni


Abstract
We present a low-rank multi-linear model for the task of solving prepositional phrase attachment ambiguity (PP task). Our model exploits tensor products of word embeddings, capturing all possible conjunctions of latent embeddings. Our results on a wide range of datasets and task settings show that tensor products are the best compositional operation and that a relatively simple multi-linear model that uses only word embeddings of lexical features can outperform more complex non-linear architectures that exploit the same information. Our proposed model gives the current best reported performance on an out-of-domain evaluation and performs competively on out-of-domain dependency parsing datasets.
Anthology ID:
W17-6305
Volume:
Proceedings of the 15th International Conference on Parsing Technologies
Month:
September
Year:
2017
Address:
Pisa, Italy
Editors:
Yusuke Miyao, Kenji Sagae
Venue:
IWPT
SIG:
SIGPARSE
Publisher:
Association for Computational Linguistics
Note:
Pages:
32–43
Language:
URL:
https://aclanthology.org/W17-6305
DOI:
Bibkey:
Cite (ACL):
Pranava Swaroop Madhyastha, Xavier Carreras, and Ariadna Quattoni. 2017. Prepositional Phrase Attachment over Word Embedding Products. In Proceedings of the 15th International Conference on Parsing Technologies, pages 32–43, Pisa, Italy. Association for Computational Linguistics.
Cite (Informal):
Prepositional Phrase Attachment over Word Embedding Products (Madhyastha et al., IWPT 2017)
Copy Citation:
PDF:
https://aclanthology.org/W17-6305.pdf