In this paper we propose a deep learning model based on graph machine learning (ie Graph Attention Convolution) and a pretrained transformer language model (ie ELECTRA). Our model was developed to detect harmful tweets about COVID-19 and was used to tackle subtask 1C (harmful tweet detection) at the CheckThat! Lab shared task organized as part of CLEF 2022. In this binary classification task, our proposed model reaches a binary F1 score (positive class label, ie harmful tweet) of 0.28 on the test set. We demonstrate that our approach outperforms the official baseline by 8% and describe our model as well as the experimental setup and results in detail. We also refer to limitations of the approach and future research directions.

Francesco Lomonaco, G.D. (2022). COURAGE at CheckThat! 2022: Harmful Tweet Detection using Graph Neural Networks and ELECTRA. In Proceedings of the Working Notes of CLEF 2022 - Conference and Labs of the Evaluation Forum (CEUR-WS.org) (pp. 573-583).

COURAGE at CheckThat! 2022: Harmful Tweet Detection using Graph Neural Networks and ELECTRA

Siino, Marco
2022-09

Abstract

In this paper we propose a deep learning model based on graph machine learning (ie Graph Attention Convolution) and a pretrained transformer language model (ie ELECTRA). Our model was developed to detect harmful tweets about COVID-19 and was used to tackle subtask 1C (harmful tweet detection) at the CheckThat! Lab shared task organized as part of CLEF 2022. In this binary classification task, our proposed model reaches a binary F1 score (positive class label, ie harmful tweet) of 0.28 on the test set. We demonstrate that our approach outperforms the official baseline by 8% and describe our model as well as the experimental setup and results in detail. We also refer to limitations of the approach and future research directions.
Settore ING-INF/05 - Sistemi Di Elaborazione Delle Informazioni
http://ceur-ws.org/Vol-3180/paper-45.pdf
Francesco Lomonaco, G.D. (2022). COURAGE at CheckThat! 2022: Harmful Tweet Detection using Graph Neural Networks and ELECTRA. In Proceedings of the Working Notes of CLEF 2022 - Conference and Labs of the Evaluation Forum (CEUR-WS.org) (pp. 573-583).
File in questo prodotto:
File Dimensione Formato  
paper-45-2.pdf

accesso aperto

Descrizione: COURAGE at CheckThat! 2022: Harmful Tweet Detection using Graph Neural Networks and ELECTRA
Tipologia: Versione Editoriale
Dimensione 1.07 MB
Formato Adobe PDF
1.07 MB 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: http://hdl.handle.net/10447/567798
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact