The recent dominance of machine learning-based natural language processing methods has fostered the culture of overemphasizing model accuracies rather than studying the reasons behind their errors. Interpretability, however, is a critical requirement for many downstream AI and NLP applications, e.g., in finance, healthcare, and autonomous driving. This study, instead of proposing any “new model”, investigates the error patterns of some widely acknowledged sentiment analysis methods in the finance domain. We discover that (1) those methods belonging to the same clusters are prone to similar error patterns, and (2) there are six types of linguistic features that are pervasive in the common errors. These findings provide important clues and practical considerations for improving sentiment analysis models for financial applications.

Xing, F., Malandri, L., Zhang, Y., Cambria, E. (2020). Financial Sentiment Analysis: An Investigation into Common Mistakes and Silver Bullets. In Proceedings of the 28th International Conference on Computational Linguistics (pp.978-987). International Committee on Computational Linguistics [10.18653/v1/2020.coling-main.85].

Financial Sentiment Analysis: An Investigation into Common Mistakes and Silver Bullets

Malandri, L
;
2020

Abstract

The recent dominance of machine learning-based natural language processing methods has fostered the culture of overemphasizing model accuracies rather than studying the reasons behind their errors. Interpretability, however, is a critical requirement for many downstream AI and NLP applications, e.g., in finance, healthcare, and autonomous driving. This study, instead of proposing any “new model”, investigates the error patterns of some widely acknowledged sentiment analysis methods in the finance domain. We discover that (1) those methods belonging to the same clusters are prone to similar error patterns, and (2) there are six types of linguistic features that are pervasive in the common errors. These findings provide important clues and practical considerations for improving sentiment analysis models for financial applications.
paper
Sentiment Analysis; Financial Sentiment; Text Mining
English
28th International Conference on Computational Linguistics, COLING 2020 - 8 December 2020 through 13 December 2020
2020
Scott, D; Bel, N; Zong, C
Proceedings of the 28th International Conference on Computational Linguistics
978-1-952148-27-9
2020
978
987
https://aclanthology.org/2020.coling-main.85
open
Xing, F., Malandri, L., Zhang, Y., Cambria, E. (2020). Financial Sentiment Analysis: An Investigation into Common Mistakes and Silver Bullets. In Proceedings of the 28th International Conference on Computational Linguistics (pp.978-987). International Committee on Computational Linguistics [10.18653/v1/2020.coling-main.85].
File in questo prodotto:
File Dimensione Formato  
10281-382981_VoR.pdf

accesso aperto

Tipologia di allegato: Publisher’s Version (Version of Record, VoR)
Licenza: Creative Commons
Dimensione 1.2 MB
Formato Adobe PDF
1.2 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: https://hdl.handle.net/10281/382981
Citazioni
  • Scopus 42
  • ???jsp.display-item.citation.isi??? ND
Social impact