City Research Online

FinSentGPT: A universal financial sentiment engine?

Mahdavi Ardekani, A., Bertz, J., Bryce, C. ORCID: 0000-0002-9856-7851 , Dowling, M. & Chen, S. (2024). FinSentGPT: A universal financial sentiment engine?. International Review of Financial Analysis, 94, article number 103291. doi: 10.1016/j.irfa.2024.103291

Abstract

We present FinSentGPT, a financial sentiment prediction model based on a fine-tuned version of the artificial intelligence language model, ChatGPT. To assess the model’s effectiveness, we analyze a sample of US media news and a multi-language dataset of European Central Bank Monetary Policy Decisions. Our findings demonstrate that FinSentGPT’s sentiment classification ability aligns well with a prominent English-language finance sentiment model, surpasses an established alternative machine learning model, and is capable of predicting sentiment across various languages. Consequently, we offer preliminary evidence that advanced large-language AI models can facilitate flexible and contextual financial sentiment determination, transcending language barriers.

Publication Type: Article
Additional Information: This is an open access article distributed under the terms of the Creative Commons CC-BY license, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Publisher Keywords: ChatGPT; large language models; financial sentiment; monetary policy; fine- tuning
Subjects: H Social Sciences > HG Finance
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments: Bayes Business School
Bayes Business School > Actuarial Science & Insurance
SWORD Depositor:
[thumbnail of 1-s2.0-S1057521924002230-main.pdf]
Preview
Text - Published Version
Available under License Creative Commons Attribution.

Download (1MB) | Preview
[thumbnail of ChatGPT_as_a_universal_economic_sentiment_engine.pdf] Text - Accepted Version
This document is not freely accessible due to copyright restrictions.
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Export

Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email

Downloads

Downloads per month over past year

View more statistics

Actions (login required)

Admin Login Admin Login