UNIOR NLP at MWSA Task - GlobaLex 2020: Siamese LSTM with Attention for Word Sense Alignment

Raffaele Manna, Giulia Speranza, Maria Pia di Buono, Johanna Monti


Abstract
In this paper we describe the system submitted to the ELEXIS Monolingual Word Sense Alignment Task. We test different systems,which are two types of LSTMs and a system based on a pretrained Bidirectional Encoder Representations from Transformers (BERT)model, to solve the task. LSTM models use fastText pre-trained word vectors features with different settings. For training the models,we did not combine external data with the dataset provided for the task. We select a sub-set of languages among the proposed ones,namely a set of Romance languages, i.e., Italian, Spanish, Portuguese, together with English and Dutch. The Siamese LSTM withattention and PoS tagging (LSTM-A) performed better than the other two systems, achieving a 5-Class Accuracy score of 0.844 in theOverall Results, ranking the first position among five teams.
Anthology ID:
2020.globalex-1.13
Volume:
Proceedings of the 2020 Globalex Workshop on Linked Lexicography
Month:
May
Year:
2020
Address:
Marseille, France
Editors:
Ilan Kernerman, Simon Krek, John P. McCrae, Jorge Gracia, Sina Ahmadi, Besim Kabashi
Venue:
GLOBALEX
SIG:
Publisher:
European Language Resources Association
Note:
Pages:
76–83
Language:
English
URL:
https://aclanthology.org/2020.globalex-1.13
DOI:
Bibkey:
Cite (ACL):
Raffaele Manna, Giulia Speranza, Maria Pia di Buono, and Johanna Monti. 2020. UNIOR NLP at MWSA Task - GlobaLex 2020: Siamese LSTM with Attention for Word Sense Alignment. In Proceedings of the 2020 Globalex Workshop on Linked Lexicography, pages 76–83, Marseille, France. European Language Resources Association.
Cite (Informal):
UNIOR NLP at MWSA Task - GlobaLex 2020: Siamese LSTM with Attention for Word Sense Alignment (Manna et al., GLOBALEX 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.globalex-1.13.pdf