Understanding differences of viewpoints across corpora is a fundamental task for computational social sciences. In this paper, we propose the Sliced Word Embedding Association Test (SWEAT), a novel statistical measure to compute the relative polarization of a topical wordset across two distributional representations. To this end, SWEAT uses two additional wordsets, deemed to have opposite valence, to represent two different poles. We validate our approach and illustrate a case study to show the usefulness of the introduced measure.

Bianchi, F., Marelli, M., Nicoli, P., Palmonari, M. (2021). SWEAT: Scoring Polarization of Topics across Different Corpora. In EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings (pp.10065-10072). Association for Computational Linguistics (ACL) [10.18653/v1/2021.emnlp-main.788].

SWEAT: Scoring Polarization of Topics across Different Corpora

Bianchi, Federico;Marelli, Marco;Palmonari, Matteo
2021

Abstract

Understanding differences of viewpoints across corpora is a fundamental task for computational social sciences. In this paper, we propose the Sliced Word Embedding Association Test (SWEAT), a novel statistical measure to compute the relative polarization of a topical wordset across two distributional representations. To this end, SWEAT uses two additional wordsets, deemed to have opposite valence, to represent two different poles. We validate our approach and illustrate a case study to show the usefulness of the introduced measure.
paper
NLP, representation learning, bias analysis
English
2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021 - 7 November 2021through 11 November 2021
2021
EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings
9781955917094
2021
10065
10072
open
Bianchi, F., Marelli, M., Nicoli, P., Palmonari, M. (2021). SWEAT: Scoring Polarization of Topics across Different Corpora. In EMNLP 2021 - 2021 Conference on Empirical Methods in Natural Language Processing, Proceedings (pp.10065-10072). Association for Computational Linguistics (ACL) [10.18653/v1/2021.emnlp-main.788].
File in questo prodotto:
File Dimensione Formato  
Bianchi-2021-emnlp-main-VoR.pdf

accesso aperto

Descrizione: Intervento a convegno
Tipologia di allegato: Publisher’s Version (Version of Record, VoR)
Dimensione 403.72 kB
Formato Adobe PDF
403.72 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10281/396231
Citazioni
  • Scopus 4
  • ???jsp.display-item.citation.isi??? 4
Social impact